June 8, '08 - Sunday - DirectX 10 and 10.1
Here is an article from Tom's Hardware (November 8, 2006) discussing DirectX 10 (DX10), which only Microsoft's Windows Vista can use. Although older, this article at [H]ardOCP gives a good description of how and why DirectX 10 works while not being overly technical in nature. In order to play a game with DirectX 10 you must have Vista, you must have a DX10 enabled graphics card and the game must be coded for DX10. Unless all 3 conditions are met the game will not run in DX10; the game will instead run in DX9. When you will actually need Vista and DirectX 10 remains a question. Only a few games have released which use DirectX10.
There has been an update to DirectX10, DirectX10.1. Note that you will need new hardware compatible with DX10.1 and a game must be coded for DX10.1 in order to use the DX10.1 code. To my knowledge, at the time of this writing (late Jan '08) no games have been announced which will use DX10.1, so it could be quite some time before you need such a card. As per DX10, you must have Vista in order to use DX10.1.
November 20, '08 - Thursday - Nvidia dedicated PhysX with mixed GPUs (PhysX chip) - Article link, Driver link
Nvidia's newest drivers add support for mixed graphic cards in order to run one as a dedicated PhysX card. While the previous drivers allowed you to run dedicated PhysX on a single card if you had matching cards, the new drivers allow for mixed cards. "How/Why is this a big deal?" You ask. Well, what this means is you can have one card, say you have an 8800 GT, and then you upgrade to a new card, say a 9800 GT. Now, instead of selling the old 8800 GT you can now use both (provided you have an SLI capable motherboard and power supply powerful enough to run both) and set the 8800 GT to be a dedicated PhysX card. You would then get the full power of the 9800 GT running just graphics and the full power of the 8800 GT running PhysX. Previously you'd be running both on the 9800 GT, theoretically losing at least some power on one application as its power is being split.
Right now there are only about 6 games on the planet that benefit from PhysX hardware acceleration, so I'm certainly not going to tell you to run right out and do something like buy an 8500 GT or 9500 GT just to run dedicated PhysX, but that is an option now. Is it necessary? Does PhysX reduce your overall framerates enough when running both graphics and PhysX on a single card to be of concern? That I can't say yet. I'm sure over the next few days and weeks we'll see reviewers asking exactly that question. I'm sure we'll also see comparisons like a single GTX 260 running both compared to something like a much lower cost option like a 9800 GT running graphics and a 9500 GT running PhysX.
There are, however, limitations. PhysX will only run on series 8 cards and higher (8xxx, 9xxx and 2xx), you will need an SLI capable motherboard, and you will need a power supply strong enough to run both cards. But this could be very exciting news as time goes on as it provides a very low cost means of users having a very powerful physics accelerator making the prospect of PhysX much more attractive to game developers as they know more people would have access to it.
At the time of this writing I'll predict that cards in the $40-75 range (new) will have more than enough power to run current PhysX acceleration. These would be; 8500 GT, 8600 GT, 9500 GT, 9600 GT. I'm betting more powerful cards would be better sold on the secondary market to recover money when you upgrade and keeping dedicated PhysX on one of these lesser cards. (It would be great if you could find one with a passive cooling system as well so it doesn't add any noise to the system. Something like this XFX 8500 GT may be an ideal choice for this.) Again, is buying an extra card a better choice than spending the money on a slightly better single card than you'd otherwise get to run both? Only time and reviews will answer that question.
June 12, '08 - Thursday - AMD causing Havok (physics) - Link
In an odd bit of news today it seems that AMD has announced a partnership with Havok for Physics. This is odd for a few reasons. First, Havok was bought out by Intel, the direct competitor to AMD, not too long ago. Prior to that Havok was working on-graphic card physics acceleration, dubbed Havok FX, in a way that both Nvidia and ATi could use. When Intel bought out Havok news of Havok FX faded away. Not too long ago Nvidia bought out Ageia, who released the PhysX physics hardware accelerator. This recent announcement of the joint effort between AMD/ATi and Havok/Intel would imply that the Nvidia PhysX would be kept proprietary, as well as the now AMD/ATi Havok physics being kept proprietary. It would be very very bad news for gamers if there were competing physics acceleration methods, as that means we could, theoretically, need a graphics card of one manufacturer and a physics hardware acceleration from the other manufacturer in order to cover both acceleration methods. (The graphic card works as both a graphics card and physics hardware accelerator.) Hopefully this will not be the case, and one manufacturer will provide the technology free of charge, and things will settle into a single acceleration tool. Splitting an already hesitant market is a very bad move in terms of future growth.
February 15, '08 - Friday - Nvidia as Ageia (PhysX chip) - Link
The dust is settling on the Ageia buyout, and it seems like what Nvidia plans to do is to tweak/incorporate the Ageia code so that certain Nvidia cards will act as both a graphics card and an Ageia PhysX chip. Those cards recently labled with "Quantum Physics", which are any of the series 8 graphics cards, will soon have a driver patch which, by my understanding, will include the Ageia PhysX code. This should allow those of us on series 8 cards to also play any game which previously required the PhysX chip for acceleration. We would then accelerate the PhysX code on our GPUs. Pretty exciting news.
February 4, '08 - Monday - Nvidia to aquire Ageia (PhysX chip) - Link
While the Ageia PhysX chip was never really embraced by the industry, it looks like Nvidia believes strongly enough in the potential to buy them out. While this is very exciting for Nvidia, it could be years before we see any real adoption of the technology or anything like on-motherboard or on-graphic card solutions. Hopefully, in the short term, this means we will see the chip mass produced in quantities large enough to get the prices down on the same kind of add-in cards we see already, most likely by companies like Evga, BFG and XFX. Apparently there will be a conference call in about a week and a half and we should know more details on the buyout and Nvidia's plans at that time.
Digital Download - Permanent - 9/22/08
First, let me start the article by saying I support the idea behind digital download of games. That being said, let it be known that I completely disagree with it's implimentation at this time. I disagree with it for a number of reasons.
The cost to the consumer is the exact same for a digital copy as a retail copy. That $55 game you pick up at the retail store is still $55 online. This makes no sense at all. In addition to the base cost, all of the stuff you get with a retail copy - manual, map, quick reference sheet, etc. - must be printed out if you want a physical copy, meaning it has now cost you more than a retail copy.
Additionally it completely destroys the secondary market. There will be no buying this version used at a reduced cost. Nope. The publisher gets 100% new price every time (provided the price is not reduced.) So, not only are they making a vastly higher profit than a retail copy, they are doing so for a longer period. As a side-effect, all people involved in the secondary market lose jobs. While it's true it has been years since I've seen a store which only dealt in used games, they are out there.
Also, you can't loan them, rent them, or take them to friend's houses. Again, due to the high costs above, this is not a good thing. Poor little Johnny who can't afford new games, but is allowed to rent on occassion, now can't game at all. (In an all digital download world.)
About the only positive thing right now that is really there ofr digital download is the lack of packaging material. Yea for sparing the environment (seriously 
). But again, this too has it's downsides. People who work in this industry would lose jobs. (Package producers, truck drivers, artists who design the packaging, and more.) And we seem to quickly dismiss the fact that the packaging and materials are recyclable and far less harmful to the environment than they once were. If we really want to worry about waste product we should examine what we are doing with food packaging. I'll bet the average gamer throws away more food packaging in a day than they would game product packaging in an entire year.
So there you have it. I support the idea behind digital download, but with it's current execution (for games) it makes no sense. It takes you less time to drive to the store and pick it up than to download it on even a broadband connection. (Unless you live somewhere remote.) It costs you just as much for far less product. It costs you more if you want a physical item. And, you can't loan it to friends, or sell / buy it on the secondary market. In my opinion I don't think digital download is worth it until they drastically lower the cost to the consumer and there are new jobs out there for those people it puts out of jobs. I like the idea and we will be ready as a society one day, but, due to (corporate) greed, I don't feel that day is today.
Game Generation - Permanent - 6/11/08
Games pretty much grow in very small, nearly imperceptible, steps. But every now and then several games will come along that really push your system and change what you need in order to play with the highest settings. I call these "game generations". Most often a generation will follow right after changes in the hardware or software industry. (Specifically a DirectX change or a change in a game engine which uses new features.) Note though when I say 'right after' this can be a period of only a few months or as long as a year.
At the time of this writing I think we are on the verge of a new game generation. (Primarily brought about by higher levels of AI, physics, and other effects.) We can see early 'growing pains' from games such as Bioshock, Crysis and Age of Conan. These games can take quite a bit more power to run at full effect levels than the games which came before. Similarly, we saw a fairly high jump in what was required to run at high levels with the change in the previous game generation from Half Life 2 to Half Life 2 Episode 1 with the addition of things like high dynamic range and bloom.
Game generations are really what will define when you need to upgrade. Hardware often times comes out quite a bit in advance of the game generation, but every now and then a game can be ahead of it's generation and there just isn't hardware out that's strong enough to run it at full effect levels. Be mindful to question if you really need that new hardware in comparison to the power required by the games you are playing. Balance the features used in current games with the features and power level of your hardware.
Riding the Wave (of Technology) vs. Leaping - Permanent - 5/25/08
Over the years I have noticed there are two different philosophies in regards to how people upgrade or replace their systems. The more common one is what I refer to as "Leaping". A "leaper" is someone who tends to fully replace their system only once in a great while. These people tend to spend a lot of money all at once, many times on a credit plan, and then stick with that same system for three, four, or more years. Someone who "Rides the Wave" is quick to adapt to changes in the industry, purchasing new technologies as they arrive, changing one or a few parts in their system, and making shifts as needed to keep current with mainstream, and sometimes high-end, technologies. A rider will make small changes to their system, often creating an ever-changing chain which makes it almost impossible to distinguish where one build ends and another truly begins.
There are two primary differences between these two philosophies. The first is financial. Obviously a leaper has a much higher bulk investment. For someone who pays all of the cost up front they are then done and don't have to worry about spending more. For someone on a payment plan they can then manage their budget around those fixed payment sizes. However, the greater difference lies in flexibility. Someone who leaps is pretty much 'stuck' with what they have for a very long time and they can't adapt to any changes that may happen in the industry. While a single change will never make their system completely useless, they can get behind in features far enough to the point that a full replacement of the system is their only option when they are ready to upgrade. Someone raiding the wave, however, makes more frequent steps, so they are much more able to adapt new technology as it arrives. If a graphic card technology comes along, a rider can just swap out the graphics. If ram needs increase, they can add more ram, if a new CPU or motherboard technology launches, they can swap out those parts.
While the financial differences can seem large, the reality of the situation is often flipped. Someone who rides the wave often is able to stay in the higher-end of mainstream, and may be perceived to be spending more money, but they can often rapidly and easily sell their old parts on a secondary market (such as ebay) and recover a lot of the cost for the new part.
Let's take a look at an example of some costs and timelines to give you a better idea of what I'm getting at. Let's say two people, a leaper and a rider, both plan on having a computer for four years total. At the end of those four years, they both expect to have a completely different system and replace everything they currently have. Things could go something like the following...
Leaper
Purchase system 1 for $3000
Wait 4 years
Replace with system 2
Total spent = $3000
Rider
Purchase system 1 for $2000
Wait 2 years
Replace graphics for $300, sell old graphics, recover $100
Replace CPU for $200, sell old CPU, recover $100
Replace motherboard for $150, sell old MB, recover $75
Replace ram for $100, sell old ram, recover $50
Wait 2 years
Replace with system 2
Total spent $2425
Note that the Rider was able to not only spend a lower total amount over the years, but they were also able to adopt any new technology that came out after those first two years. The Leaper had to spend a lot more up-front, as they knew the system would have to be strong enough to hold them for four years. Additionally, the leaper was 'stuck' with any technology they had, as they didn't plan to make any changes to the system. Thus, any new technology that games require would not run on the leaper's system.
While it is easy to see that riding the wave is the better choice, it often times isn't as simple as changing how you think about computes and your upgrade habits. Someone who rides the wave must have the knowledge to change the parts in question, as well as having the time and understanding to do so. (Or know someone with the time and knowledge.) Additionally, since they have to deal with selling their older parts, they either have to have a decent knowledge of how to do that, or a reliable person (such as a friend or family member) who would be willing to buy their old parts.
I would recommend that everyone ride the wave. You can stay current with all your technologies, and if done carefully, you can always keep your system in the higher end of mainstream as you can adopt new technologies right when they drop out of the bleeding edge range. However, I acknowledge that most people simply don't have the time or knowledge, and leaping is the better choice for them. Take a moment though and consider your own choices. Figure out if leaping or riding is better for you.
Planned Obsolescence vs. Future Proofing - Permanent - 5/23/08
Once upon a time, people looked to "future proof" their system. What this meant was that they would adopt the most bleeding edge technologies that they (and often the industry) believed would be adopted in the future. I myself was one of these people back in the day. (I still am sometimes.) However, I feel that the advancement rate of technology, particularly in the computer industry, has become too fast to consider this an ideal choice. Instead, I now feel that "planning for obsolescence" is the better way of looking at your hardware.
Consider the thought behind "future proofing" to be that you adopt the technology early, paying a high price to adopt it while it is bleeding edge, as a way of sparing yourself from needing to make the change later on once the technology is mainstream. Consider that "planning for obsolescence" is choosing the technology which is proven and mainstream, while keeping your eyes open and keeping yourself ready to make the move to the next technology once that technology has evolved and proven itself.
What happens is this: a bleeding edge technology will launch, and it will do so in a non-perfect condition, and at a very high price. With this bleeding edge technology you get cut in a few different ways. Mostly you are killing yourself on price. Typically price for these technologies is ridiculously high compared to the mainstream technology. But you are also hurting yourself on performance. While this tech may be the next step in terms of evolution, the problem with the bleeding edge parts is that they will rapidly be replaced by something better as the technology evolves and it floats into mainstream. More often than not, by the time the technology has reached the upper end of mainstream consumption there are better and cheaper parts available.
In my opinion you are really doing yourself a disfavor by adopting bleeding edge in an attempt to future proof your system. Future proofing, in the way it used to exist, just doesn't happen anymore. (If it ever did actually exist to begin with.) With the frantic pace we currently have in the computer hardware industry pretty much every part you have will be completely out-done in terms of performance and features within two, three years at the most. It is far better in my opinion to wait until the technology has floated down to the upper-mainstream segment, and then adopt it at that time. Adopting a technology early won't prevent the fact that when that next technology comes you will have to change. Think of it like eating meals. Sure, you can eat early, but that won't prevent the fact that, at some point, you are going to need that next meal.
Sure, someone has to buy the bleeding edge parts. Someone has to be the early adopter. But unless you have so much money it is effectively meaningless to you, I really recommend you buy your parts with "planned obsolescence" in mind rather than trying to "future proof" your system. Instead of saying, "here, let me pay triple the price for this one part that will be really great in a year", say to yourself, "hum, that part is triple the price, but it won't be a benefit to me for another year. By then there will be faster versions of it available at a much more affordable price. I'll wait and adopt it at that time." Technology is rapidly outpacing need. And the simple fact of the matter is, particularly with gaming, you simply won't have an advantage by having that bleeding edge part. It just won't give you any kind of gain because games aren't designed to take advantage of what it has to offer. It takes years to develop games, and by that time, said bleeding edge part will have floated into the mainstream segment and be far more affordable.
Social Hubs - Permanent - 5/16/08
I believe social hubs are critical for online gaming. While I certainly didn't invent the term I'll give a brief description. I define a social hub as a physical location - though I guess in some cases it may also be one not physical, such as a chat room - where people naturally gather.
Recent online games seem to lack these social hubs, which do seem present in the online games that have been more successful or more popular. I feel it is these social hubs which allows players to feel connected, like a part of a greater whole, and that feeling keeps players far more interested in the game, and far more socially active. These can be designer created, as well as ones that naturally form due to the actions, and interactions, of the players. Examples of developer created social hubs would be where they place crafters or trainers, how close these NPCs are to one another, how the pattern of traffic flows between areas and between the NPCs occur. (Think of Ironforge in World of Warcraft.) An example of a player designed hub, or one which naturally occurs, are places such as zone lines, or places where players are commonly waiting, such as a public transport system, or returning to, such as an area with quest givers or vendors.
I have thought about all the games I've played since Everquest (EQ) and I have come to the conclusion that these social hubs are really what has kept me social in games, and that social connectivity between other players is what keeps me interested in the game itself. I played Everquest for roughly 1.5 years (starting at launch), the longest of any online game I've played so far. The second longest game so far was World of Warcraft, which at the time of this writing I played from launch until 6 months later. (I say at the time of this writing as I am considering returning to check out the expansion.) Everquest was filled with social hubs. There were zone lines, players would wait at zones for other players and cross together if the zone was dangerous. There were travel points, where players would gather in groups, or wait for transport. There were places where players would gather looking for groups, or dropping off members who wanted to leave a group after they'd been to a certain dungeon. All of these places created a spot where you not only saw other players, but interacted with them, either directly or indirectly. It created a connection between people where you could see other people you knew, and other players that you may not normally interact with. Like back in the day with EQ you really didn't see more than one person of one class at a time, so, for example, since I was a Cleric, I never really grouped with other Clerics, so I would only see my fellow Clerics in passing. But, due to these social hubs, I did see them, and there was that camaraderie between us.
Recent online games seem to be forgetting to include these social hubs. Some worlds are focusing on becoming so "massive" that players are spread so thin that these hubs simply won't ever form, or they aren't designing the game with these kinds of hubs in mind, so players aren't encouraged to travel and pass one another on connecting routes. Other online games are going the opposite of "massive" and are using a heavily instanced, or solo play style. This is equally as hubless as it doesn't allow for any interaction between players.
For me at least, when I look back at the many online games I've played over the years, the ones which I found the most fun, and that I played the longest, had some form of solid social hub system. Other games, ones I wound up leaving fairly quickly, lacked strong social hubs, and in some cases lacked them altogether. I really think a good hub system is critical for an online game's success. You may want to reflect upon your own experience and ponder the social connectivity you've seen in games. It's possible you too have been draw into, or away from, a game due to it's inclusion, or lack, of social hubs and player interaction.
