"High-end" graphics has nothing to do with colour depth, per se. I think that it has much more to do with:

(1) the "photo-realism" effect that those games were aiming for (as opposed to the "fantasy" setting of WoW); for better or worse, Age of Conan was aiming to be "dark and gritty", for instance; and,

(2) laziness on the part of the developers. For example, I think that both Age of Conan's and Crysis' opening levels were fairly colourful, but the colour tended to wash away as the game wore on, suggesting that the graphics artists spent more time at the beginning than the ending.

(In contrast, I was fairly impressed by a recent Star Wars: The Old Republic walkthrough video which showed a really vibrant, interesting universe.)

You bring up two interesting points:

- WoW's graphics look dramatically out-of-date (after all, the game was released in 2004, with comparatively minor updates done over the years, and the polygon counts are simply too low)
- zones appear to be completely unable to handle a load of, say, more than ~40 players actively engaged in close-quarters PvP combat (witness the 'Wintergrasp' zone)

  That said, it is unlikely (in my opinion) that the lag you experience in WoW when there are many players in a zone has much to do with the graphics engine, and probably has much more to do with Blizzard's "backend". From what I have personally experienced as a WoW player (and as an experienced software developer), I suspect that their database server isn't able to keep up with the massive number of queries/inserts/deletes that are required during active PvP. This doesn't necessarily point to a problem with the DB engine itself -- it usually arises because of poor database layout (there is a certain "art" to database design, balancing normalization vs. redundancy to reduce query frequencies, setting up indices, and so forth), poor SQL (it never ceases to amaze me the kind of queries that some people write), and poor overall use of the system (meaning, the server is simply processing so much information that it gets bogged down). I think that this is a fundamental issue with the WoW server/backend which has existed in the game from its inception, but which had been cleverly hidden in most regular play scenarios (e.g., generally by imposing a 40-man cap in the instanced dungeons/battlegrounds/etc.).

Given my conjecturing, I didn't find it at all surprising that Blizzard has a job open specifically focused on "performance implications of database architecture choices". But of course, this is all an opinion & speculation, so who really knows.

MS Word isn't "more widely used than TeX" in the academic community.

In a typical IEEE or ACM journal or conference, I would estimate that easily 98% of the papers are written in LaTeX (judging by the fonts and the layouts).

And, at least in engineering, most Master's or Doctoral theses are written in LaTeX.

The reasons for this are simple:

(1) BibTeX offers better bibliography formatting & support than anything provided by MS Word.

(2) When used in combination with a source code control system like CVS, LaTeX simplifies collaborative document writing. (e.g., multiple researchers can work on the same .tex file and simply have CVS merge changes together.)

(3) Writing complex documents -- such as a doctoral thesis -- where one must have a ToC, LoF, LoT, and multiple appendices each with their own bibliography can require a substantial amount of time to format in MS Word. (You'll be inserting section breaks, worrying about running vs. non-continuous headers & footers, page number consistency, and so forth.) On the other hand, these tasks are largely taken care of for you by existing LaTeX templates.

(4) Various advantages when representing complex equations.

As a Waterloo grad student, I can say for sure that Waterloo's relaxed IP policies help in securing funding from industry.... (Companies are more likely to fund research if they can benefit from it directly, and they can do so more easily when the researchers that they fund are not hindered by draconian IP regulations.)

> most people are not even knowledgeable enough
> about their computers to even understand what
> free software is all about, why it matters, and
> why they should care.

To add to this, I think that there are many people who are familiar with free software, but who do not want to go to Stallman's extent of refusing to use or interact with non-free software.

Personally, I view software like I view any other tool in my workshop: I have some tools that I've made myself (on a lathe and all), I have some tools that I was given for free, and I have some tools that I went to Home Depot and bought outright. I use each of them, in different ways, for different tasks in order to maximize my overall efficiency and minimize my overhead.

In much the same way, I use free software for some tasks and commercial software for others. To blindly commit myself to using either free or non-free software would severely impact my productivity.

(I think that there are a lot of people who employ a similar, "moderate" philosophy.)