Google Wave

I’m in bed with some kind of virus and was thus sufficiently bored to actually watch the entire Google Wave Demo. It’s a very impressive demo and a very nice piece of technology, and it’s being implemented in an idealistic fashion.

The big problem with it is that it isn’t backwards compatible with email. I would have thought that being able to interoperate with email would have been a major priority, but in fact the only mention of it was a half-mumbled remark towards the end about how it would be nice if someone wrote an extension for this. It’s especially odd that this issue didn’t get dealt with more explicitly, since Wave is explicitly intended to replace email.

It’s interesting to imagine what a conversation between a Wave user and a conventional email user might look like — either the email user is going to lose all context each time he/she gets a reply, or the Wave user is going to get a distinctly un-Wave-like mess of pointlessly quoted crap.

Star Trek

There are “trekkies” — people who love Star Trek despite the fact that it’s mostly just terrible, and “disappointed trekkies” — people who love what they imagine Star Trek could be if it weren’t just terrible. What if, for example, we considered life in the Federation for people not working in Star Fleet? (In the new Star Trek movie, when Earth is attacked (seriously that isn’t a spoiler, right!), the only reaction we see is of Star Fleet Academy cadets running around in terror. Some things never change.)

For example, consider that “godlike omniscient beings” and “time travel” are generally considered the last resort of the desperate, or — at best — a good excuse for comedy, in Science Fiction. Yet, these two devices are associated with much of the best in Star Trek (e.g. most people like the Q stories in TNG and cite “City on the Edge of Forever” and “Yesterday’s Enterprise” as among their favorite TOS and TNG episodes.

No doubt my reader has guessed that I’m a “disappointed trekkie”, but I count myself among those who enjoyed the new J.J. Abrams Star Trek (XI) movie. I’m not sure I approve of it, though, since it carelessly used the interstellar transporter technology (OK, a slight spoiler) from the lamentable Spock Must Die to resolve a stupid plot point that shouldn’t have been there in the first place.

So, a well-made, fun movie, but we’ll all have to pretend that bit never happened the way we all eject our Wrath of Khan DVDs before Kirk’s eulogy starts.

Fallout 3

I’ve just finished Fallout 3, most of it in “easy” mode, and I thought I’d share my impressions.

I was a huge fan of Fallout and Fallout 2, and both were among the few CRPGs I’ve completed more than once (something they encourage by actually allowing you to make decisions that have consequences). Much has been made of the fact that decisions in Fallout 3 have consequences, and the fact that — say — insulting someone in a conversation, or killing an NPC who is also a quest giver, might have actual consequences is not only unusual today, but it was unusual back when its predecessors came out.

The Cost of Consequences

This isn’t innovative though, quite the reverse. When it shipped, Fallout was a 3d tile-based (indeed hexgrid-based) RPG of epic scope with 100% 3D rendered graphics. The amount of work that must have gone into producing all those images is simply mind-boggling. Producing actual 3d models is much easier and more forgiving that building animated pre-rendered figures.

Anyway, the cost of all this graphics production is generally linear game-design. Back when games were text-based or simply re-used a small library of graphics over and over again it was relatively inexpensive for player decisions to have consequences and so it was also relatively common. As the costs of game production have gone up (and most of this has been graphics production — music and audio haven’t become substantially more expensive to produce the way graphics have) the desire of game developers to spend money on stuff most players will never see has gone down.


Fallout 3 is graphically very impressive, astonishingly detailed and textured, but also very monotonous. Its color palette is FPS brown and grey, and pretty much every building interior looks identical. You basically spend a lot of time outdoors in largely unvegetated rubble and rock, or in collapsed subway tunnels, ruined office buildings (with huge numbers of toilets), and the occasional house or sewer. That’s it. You see the same metal box, locker, desk, chair, bus, wrecked car, over and over and over. Or so it seems. It doesn’t look like copy-and-paste, but there is a huge overall sameness to most of the world.

It bugs me that even the people who are supposedly living quite well seem to put up with their houses looking like shitholes. (I’m ignoring the final bad guys, who live in near Star Trek utopia, although even they have radioactive cockroaches scuttling around under their floors.) This is all pretty consistent with the original Fallout, but it does get boring after a while.

Let me get this straight: not only can I kill a quest-giver, but the quest might be able to cope with it?

Although I loved Arena and wanted to love Daggerfall, I was not impressed by Morrowind and skipped Oblivion, so I’m not sure how much of Fallout 3 is simply “post-apocalypse-flavored-Oblivion”, but there are some aspects of Fallout 3 that border on genius. To begin with, “quests” are infernally complicated, you often don’t get enough information to really finish them straight away, they “morph” as you do them (e.g. discover that something you intended to do can no longer be done, or never could be done), and they are all pretty intricately woven into the game world.

Here’s a short example with (minor) spoilers. When I first entered the central hub town (“Megaton”) I met the sheriff (it’s hard to avoid). He gives me something of a run-down on things and I learn there’s an unexploded atom bomb in the middle of the town, and suggest I might be able to defuse it. Later (I have no clue how to defuse it) I bump into a “mysterious stranger” who offers to pay me well to set off the bomb (after a suitable delay to get away, one assumes). He seems kind of scary, so I agree, planning not to do it but figuring I’ll play along.

Later, I’m talking to the sheriff about something else and notice a new conversation option “hey this weird guy offered to pay me to blow up the town, and he gave me this”. I pick it and the sheriff decides he wants to talk to the guy. So we go to the saloon and the guy basically denies everything and gets up to leave. The sheriff tries to arrest him and the guy pulls a gun and kills the sheriff. I had been standing back a ways because as the conversation progressed it sounded like the guy was going to accuse me of being the terrorist (or whatever) and figured I might end up in a fight, so I pull a gun on the guy who suggests to me that I’d be better off getting out of his way. I kill him.

I then get notified by the game engine that I can no longer collect on the “blow up the town” quest (since I killed the turn-in NPC, I guess). Personally, I think I could have figured that out for myself, but maybe they didn’t want people to complain there was a bug in the quest (after all World of Warcraft players will not be able to cope with the concept of quest-givers being killable and/or not respawning). I loot both corpses (no-one in the town seems to mind, but note that they will get mad if you kill someone they like or steal their stuff) and find the sheriff’s key. Figuring he doesn’t need (or own) his stuff any more and I’m desperate (I barely have ammo for my half-broken pistol) I let myself in his house. Everything is still flagged “property of” someone or other (which I assume is a bug, but turns out not to be…), so if I take anything it will be bad karma (literally, the game tracks your karma, which is neat).

So I figure I’ll only steal guns or ammo. After I pocket a couple of choice items, a kid’s voice asks me something (I don’t remember what), but it’s the sheriff’s kid and he just found out his dad died. (And here I am with pockets full of his dad’s stuff and, heck, I got him killed.) D’oh! At the end of the conversation the kid mentions his dad left him a bunch of stuff for me in the event I ever defused the bomb…

I won’t go on, but this is literally a “tech level above” World or Warcraft or Knights of the Old Republic in terms of quest construction, even if it’s still traditional, “static scripted” content. Now, the world itself does not change dynamically the way the worlds of Fallout and Fallout 2 did (you could literally overthrow governments and come back to a different society) but it’s a far more interesting world than any I’ve seen in any kind of RPG for a long time. Bravo.

Fallout 3 is not without its flaws, of course. You’ll spend a lot of time in combat, and combat kind of sucks.


The combat system is a hybrid of Fallout’s action-point-based tactical game (where you could opt to shoot at a target in general or a specific body part) and a first person shooter, but the weird thing is you have action points that you use to fire at specific body parts which are entirely separate from the FPS. Thus, shooting at someone’s head using the “VATS” (targeting system) costs you action points while just pointing your crosshairs at someone’s head and pulling the trigger doesn’t. I think I’d been playing the game for about twenty hours before I even tried to shoot someone while NOT using the VATS. Sadly, the chief reason to do so is that you often run out of “action points” to use the VATS, but VATS saves you a lot of ammo. Typically (and this is really stupid) I’d use VATS to headshot a target initially, but free fire to finish off near dead targets (to save action points). The idea of shooting someone in the head three times with a sniper rifle and then having to finish them off is both really stupid and a betrayal of Fallout’s legacy. (The main reason I switched to “easy” mode was I got into a fight with “super mutants” on hard which I simply couldn’t win… super mutant leaders can take two or three headshots in “very easy” mode and not die.)

Enemy AIs are horrible. On the one hand, enemies don’t tend to do anything very interesting, and on the other, they have perfect reflexes (e.g. if you shoot someone in the back of the head while concealed they instantly turn around and shoot you).

As alluded to in the preceding paragraph, the way wounds work is stupid. While the game keeps track of where you get hit, and tells you if you’re crippled in a limb (if your legs are crippled, you can’t move quite as fast, if your arms are crippled you can’t shoot quite as accurately) wounds have almost no visible effect on enemies. Many many times I would shoot someone in the head, get a critical, and they would continue to fire at me without even a pause. (Crippling a target’s head seems to cause them to rub their head for about 2s.)

Finally, the way cover works (or, rather, doesn’t) is incredibly annoying. You can’t “lean around” corners, and there’s no explicit cover support. Sometimes you’ll be able to see a target’s head but the engine will tell you it’s impossible to hit. Worse, sometimes you’ll take a bead with a 95% hit probability and then the engine will shoot a rock or girder for you. Some weapons consistently hit low objects (e.g. the minigun simply won’t fire past anything more than knee-high, making it virtually useless).


The Fallout games are probably funnier in retrospect than they were in actual play, but Fallout 3 is pretty dour compared to either of its predecessors. Probably the funniest thing in the game is the Enclave’s presidential radio broadcasts which are a mixture of insane right-wing bullshit and social commentary that seems a little bit more about our world than Fallout’s. A lot of the humor in the original games concerned the various oddball societies you encountered, but in Fallout 3 you don’t really encounter many different societies until close to the end (when you encounter two fairly colorless high-tech civilizations).

Final Thoughts

Don’t expect a feel-good ending. Like its predecessors, Fallout ends on a downer. (I haven’t played through the different possible endings, but as far as I can tell I ended the game about as positively as possible and it was pretty downbeat.) That said, I really liked the ending and it was satisfying. I’m not even sure I’m going to backtrack and play through more bits I missed or play the game through again, but then it took a pretty substantial chunk of time (it took me 54h according to the machine — probably including about 24h of pausing). While Fallout 3 feels like a smaller and less ambitious game than either of its predecessors, it still stands as a good story in its own right.

Fallout 3 is, in terms of its “30s cycle”* a shooter, and as a shooter it’s pretty broken. I think Grand Theft Auto IV’s combat, which explicitly supports cover, is far superior to Fallout 3’s, which is sad because GTA IV is a driving game with a shooter component, while Fallout is shooter with no other components (well, unless you count scrounging for ammo and inventory management). Add VATS to GTA IV’s shooter component and you’d have a pretty nice game. (I’ve yet to play Gears of War, but from what I’ve seen and read this is the “FPS” Bethesda should have ripped off when they tried to turn Oblivion into a shooter.) That aside, AIs which have perfect aim and instant reactions is pretty lame for a modern game; it doesn’t make a game hard, it makes a game stupid. Similarly, having to shoot people three times in the head before they even slow down is ridiculous; more ridiculous when it’s on “easy” level. I don’t know if my nostalgia for Fallout and Fallout 2 is clouding my memories, but I seem to recall it actually had NPC morale. In Fallout 3 you can be attacked by four guys, and blow the heads of three of them with three shots and the fourth will still run towards you like an idiot.

So, in a nutshell: loved the story, loved the quest system, but the look of the game and the combat system eventually got old.

* Quoting the designer of Gears of War in a recent New Yorker article.

Quick Guide to Fallout 3

You play your character — literally — from birth. Nice idea but REALLY annoying for repeat play. The childhood bits are tedious, you can’t “click through” them, and they have no real effect; I’d much rather have the option to just create a character how I wanted it (which you get to do at the end of your childhood anyway).

“SPECIAL” (Strength, Perception, Endurance…) attributes. Some folks say “pick strength because it lets you carry more stuff”. They have a point but Strength 1 = 160lbs, 10 = 250lbs. It’s not a huge deal — you’ll constantly be fast-traveling to merchants regardless. Typically you “need” about 80-120lbs of stuff. I carried a lot more because I never knew what would be useful. Turns out nothing much was useful and I could have gotten by with much less. Perception’s main effect is determining the range of your “radar” (how far out your red dots show up). The red dots aren’t that useful, but having really low perception probably sucks. Endurance affects hit points. Again, having really low endurance probably sucks since you can avoid wasting stim packs if you get get to a (free) bed to recover health. Charisma is important — some dialog choices won’t be available if you’re ugly. Intelligence gives you skill points, but the effect isn’t that marked. Agility drives APs for your VATS (targeting system). Luck affects everything and especially criticals. I had Luck 5 and can’t really tell how important it is (I did get a LOT of criticals, but I took any perk that helped with crits).

Edit: I checked and INT does not affect skill points received nearly as much as I thought so my advice was garbage. So…

Here’s my recommendation for allocating attributes: Agility gives you more points in VATS, so if you like blowing peoples’ heads off, get lots. You might not want 10 points because when/if you find the Agility bobblehead (which gives you +1) you won’t benefit if you have 10. So you might do something subtle like everything 5 except agility 10, or maybe agility 9 and perception 6. On the whole, stats don’t make a huge difference.

Repair skill is interesting. Essentially it represents cannibalizing parts. This allows you to take two pistols in lousy condition and turn them into one pistol in pretty good condition that is also worth more, and as such it is like having the ability to “carry more” since you can concentrate value in a smaller number of items.

Initially you want one weapon skill only, and that weapon skill is small guns. Do not waste skill points on more than one weapon skill. The choices are unarmed, melee, small weapons (pistols, rifles, shotguns, SMGs), energy weapons (lasers and plasma weapons but not the gatling laser), and big guns (missile launchers, miniguns, etc.). I haven’t tried unarmed and melee — they may work really well, but given the number of opponents with ranged weapons expect to spend a LOT of time hiding around corners or sneaking up close to enemies. Given that you spend quite a bit of time being attacked by multiple, spread-out enemies with guns, being a melee specialist seems like a world of hurt to me. Get your main weapon skill to 95 fairly quickly (not 100, since you’ll find books to increase it for “free” and they won’t work if you’re at 100 already). Eventually you’ll want big guns (for the missile launcher, flamer, and gatling laser), but there’s a perk you can take three times that gives you 15 points of big guns and missiles don’t really need to be aimed.

Why I recommend small weapons

First, you’ll have almost no choice in the early game.

Second, they’re ranged, which means you can often take out targets from a distance while sneaking (a huge benefit).

Third, until you get big guns (the missile launcher in particular) you won’t find anything more effective than the Chinese Assault Rifle or Sniper Rifle (I haven’t experimented with the home-made weapons yet, maybe some of them are ridiculously effective) until practically the end of the game.

I might add that I am a pack rat in every RPG I play. I didn’t sell a single round of ammo in my first play-through and I think I found a total of less than 30 missiles in total. The minigun is terrible (it doesn’t work from behind any kind of cover and takes ages to fire, during which you’re getting pounded), and other big guns have ammo even rarer than missiles.

Also note that the missile launcher and flamer don’t seem to require any real skill to use (you don’t tend to aim them using VATS) and the energy weapons (laser- and plasma- weapons) seem to be ridiculously accurate even with low skill but to do no real damage. Many a time I’ve blown off the head of an Enclave Soldier in power armor while he ineffectively plinked at me with a plasma rifle.

Weapon DMG ratings are highly deceptive. The damage is “per shot” so the weapons that fire bursts effectively (a) have a much higher chance of scoring a hit in the VATS system and (b) do a ton more damage. That said, the minigun is slow to spin up and down and won’t fire over knee-high obstacles, so it still sucks. The assault rifle and SMG are your friends (but use VATS or you will run out of ammo very fast). Eventually I got a special (better) SMG but only because I screwed up and let an NPC I liked die. (And I was not restoring from backup unless I died.) Later on, the flamer is a very, very effective weapon (but only at close quarters).

You need to be good at one of science and lockpick. Science is used pretty much exclusively to hack computers. Usually anything you need to get into can be hacked or lockpicked into. Also note that you get quite nice XP awards for picking a lock or hacking a computer. The hacking subgame is very cute, reminiscent of “Mastermind”. The lockpicking subgame requires a small amount of dexterity (a light touch, mainly) and bobby pins, but is quicker. If you’re easily bored or frustrated but dextrous, take lockpicking.

That’s about it. You need Repair, Small Guns, and Science or Lockpicking as your “tag” skills, and when they’re maxed you can experiment with others.

Of the other skills, medicine is not as useful as I hoped. It makes stim-packs and other “healing potions” more effective, but I ended the game with a huge stockpile of both money and stim packs. I might have needed more on a higher difficulty level, but I also never “restored from backup” unless I died (which happened in two places). Speech is amazingly useful — it gives you more dialog options which often allow very interesting plot options / bypasses. You also get XP for using your Speaking skill.

Sneak may be very useful, but I haven’t really tried. I wasn’t interested in picking pockets, and despite spending almost no skill points on sneak I had little trouble sneaking into sniper rifle range in open areas, while indoor fights were mostly “corner-jutsu”. I had a ton of “Stealthboys” in my inventory (among other things they give you +100 sneak, sort of like an “invisibility potion”), but the one time I used one I found it almost useless, so I’m inclined to think sneak is not very useful for combat, so for thieves only.

Barter will make you richer, if you care. But repair will make you more money and is useful for other things.

So the upshot is small guns, lockpick / science, repair are you most important skills, with speech next in line, and barter and medicine bringing up the rear. You can raise big guns in a hurry late in the game when you start actually getting some big guns, and energy weapons, melee, and unarmed are all skippable.

Fallout: New Vegas note: small guns is now just guns and is still the skill of choice in my opinion, although sneak seems more useful and is apparently fun in combination with melee / unarmed; lockpick and science are not as interchangeable as they were in F3. Science is quite useful on its own (and in many quests it lets you bypass or simplify a medicine or repair task), but lockpick is indispensable (unless you plan to simply force locks).

I think it would be very interesting to play either game with a character who tags speech, sneak, and medicine and try to finish the game without killing anyone (directly!).

Another Annoying D90 “Feature”

Turns out that the D90 uses lossy compression on its RAW files.

The D90 achieves relatively small RAW file sizes by applying lossy compression to the data whether you like it or not – to put them into context, the D300 with the same resolution typically delivers 12-bit RAW files measuring 13.6MB with lossless compression, or 14.2MB with no compression at all. It’s always good to save space, but we’d sooner the D90 employed lossless compression on its RAW files, or at least gave you the option. For that you’ll need the D300. (

This is a really annoying decision. It’s bad enough that the D90 doesn’t offer 14-bit image processing (something you get with the much cheaper Rebel XSi, and the slightly cheaper 40D) but this is just dumb. Perhaps Nikon’s lossy compression is so damn good that no-one can tell the difference, but why not make it an option. I guarantee not having a lossless file format will cost Nikon sales.

Deliberately crippling cameras (heck, I’m betting that Nikon is using the same circuitry as the D300 and just flipping some firmware switch to disable 14-bit processing and uncompressed RAW) means that the digital camera market is teetering on the edge of massive price drops (after all, it’s clear that the D90 is a D300 in a flimsier box for 60% of the price, or would be sans deliberate crippling), which are all the more likely given the economic situation.

It used to be typical for camera companies to sell a ruggedized version of a consumer camera for a ridiculous amount of money. For that you got interchangeable components (focusing screens, backs, viewfinders), weather seals, longer life components, faster motor drives, and so forth). But the cheaper camera didn’t have a special circuit in it to deliberately fuck up your pictures.

Until Nikon is willing to stop insulting my intelligence, I’m not buying another piece of Nikon kit.

The Nikon D90’s Annoying Flaw

Nikon’s long-rumored and just-announced D90 is a pretty wonderful piece of kit by all accounts, but with one notable and annoying limitation buried in its tech specs. It has a 12-bit A/D converter whereas Canon offers 14-bit support in its similarly priced 40D (indeed I can find 40D’s selling from reputable vendors for under $950) and considerably cheaper EOS 450D. While providing HD video capabilities in the D90 is certainly a radical, bold, and perhaps even compelling move — skimping on still picture quality (tonal* range) is pretty annoying.

Perhaps Nikon is trying to differentiate its “pro-ish” D200 and D300 from it’s “enthusiast” D80 and D90 with tonal* range. If so, this seems to me to be a grave misjudgment. Serious photographers will be intrigued by the video shooting capability, but ultimately I think that — effectively — limiting the D90 to inferior quality “film” is very, very bad idea. Nikon never did similarly dumb things with film cameras — why do it with digital?

Nikon is now in the rather odd position of having a huge gap between the D60 (which retails for around $600) and the D200 (which retails for around $1300) which has two Canons (the 450D and 40D) in it, both of which shoot 14-bit RAW.

Now, just how big a difference 14-bit makes over 12-bit in practice is unclear. Pixel-peeping review sites such as haven’t really started looking hard at it, but it’s obviously important enough that both Canon and Nikon have added support for it in some of their cameras. In theory, each “bit” should equate to an f-stop of dynamic range, which would make it ridiculously important.

Anyway, leaving 14-bit RAW support out of the D90 seems like a foolish oversight. It gives people a reason not to buy the D90. The D80 simply did everything it could reasonably be expected to do and then some. It was incredibly successful because there was simply no reason not to buy it. The D90’s lack of 14-bit support will give me, for one, pause. Of course, Nikon may release a D90x with 14-bit RAW once it ramps up production … in six months.

Second Thoughts

If you visit the Nikon’s D90 web page you can see a sample of the video quality, which is stunning. Autofocus is disabled while shooting, so it’s not going to be an all-round camcorder substitute, even with a separate audio recorder (which is a good idea even if you’re using a real camcorder) and an external battery grip. So, video is definitely not a half-arsed feature.

While the 12- vs. 14- bit issue is galling, it’s worth noting that the D90 excels the 450D (and in several cases the 40D and 50D) in some rather more important ways, such as continuous shooting speed (4.5 vs. 3.5 fps), 3D predictive autofocus, autofocus points, metering, and viewfinder quality. So, while the lack of 14-bit NEF support is still grating (reminiscent of Apple blocking dual display support from iBooks to differentiate them from Powerbooks way back when) Nikon is delivering an all-round superior camera.

The street price of the D300 is dropping, too, so I guess Nikon is bracketing the 50D with the D90 and D300, and letting the D90’s superior usability deal with the 40D and 450D.


As per comments from an alert reader, 14-bit refers to tonal range not “dynamic”. The two aren’t completely independent (as per my responses in comments) but I stand corrected.

Post Script

dpreview has posted sample shots from the Nikon D90 and also the Canon 50D. The first thing I look for is good low light performance, and both produce very usable output even at ISO 6400 (if you pixel peep, it looks like the Nikon produces “clusters” of noise while the Canon produces more outright artifacts, but this is just one photograph per camera of different scenes).

There’s no question that the $1000 D90 trounces the $800 450D. Although the Canon has 14-bit RAW, it is built like a toy, has a pentamirror* viewfinder, maxes out at ISO 1600, uses its flash for low-light focusing, and offers lower shooting speed (3.5fps vs 4.5fps). The 50D barely trumps the D90 for considerably more money with compact flash, 14-bit RAW, and slightly higher resolution… but the D300’s price is now in the same ballpark as the 50D.

Canon has finally released an image-stabilized 18-200mm lens (quite reasonably priced too) which will at least remove that as a reason to go Nikon.

* Every film SLR I’ve ever owned or used has had a pentaprism with excellent coverage, and not one cost more than $350, so why on earth do we find pentamirrors acceptable in $800 digitals?

Canon 50D at 6400 ISO
Canon 50D at 6400 ISO
Nikon 90D at 6400 ISO
Nikon D90 at 6400 ISO