Real Innovation is a Change in Usage Behavior

One of the problems inherent in the PC/CE/tech industry is that people get entrenched in specific ways of thinking, especially on the hardware side. At this point, the vast majority of hardware is commoditized, standardized, and off-the-shelf, so companies that fight over those markets start to think about chassis, overclocking, and paintjobs as competitive advantages. Sure, these things enable you to achieve a unique look or better cooling/performance, but they don't actually provide users with dramatically different ways of using products. It's not innovation if you take a product and make it look different -- it's still a PC or a smartphone. Innovation is when usage behavior changes.

Take the litl webbook, for example, which was just launched a few days ago. It's a laptop, but you won't use it like any other laptop or even a netbook. After all, there is no hard drive, no operating system, you can't install software, and you can only use it within the range of wifi. That certainly cuts out a huge swath of activities that you'd normally do with a PC, so it's not actually a PC and doesn't replace the one you have. Instead, its main interface is a simple to use remote control, it folds into an A-frame so you can easily set it up like a TV (the display is also extra-bright with a very wide viewing angle), and you'll never touch a device driver, virus utility, or system setting because the software is all in the cloud. This is a machine that will find its way into the hands of people that would never touch a traditional computer, and into places your PC just doesn't belong, just because of its sheer simplicity and ease of use.

You can tell something's different when you describe a machine not by hardware specs, but by who, how, and where it will be used.

If you take this one step further, you bump into Bug Labs. This is a modular, open source hardware platform that dares the user to innovate for themselves. While this product in its current form is nowhere near ready for the general consumer, I think the IT solution provider, tech enthusiast, and gadget entrepreneur will find this liberating. Instead of hardware defining what's possible, the BUG provides an incredibly economic way to build the exact device to fit your usage. It provides a completely new way to think about and use technology, not defined by specs, but by what you want to do with it.

Thinking about innovation like this is what creates an OS like Chrome, an mp3 player like the iPod, and a video game system like the Wii. And innovation is never-ending, even in a ridiculously stagnant market like the PC market.

Continue reading >>>

Smaller Business Doesn't Automatically Equal More Agile Business

Lou Gerstner, Chairmain/CEO of IBM from 1993-2002, brought the company back from a dying has-been to a relevant, powerful force which helped to usher in the internet age.

In his memoir "Who Says Elephants Can't Dance" (2002), Gerstner writes:

"For much of my business career, it has been dogma that small is beautiful and big is bad. The prevailing wisdom has been that small companies are fast, entrepreneurial, responsive, and effective. Large companies are slow, bureaucratic, unresponsive, and ineffective.

This is pure nonsense. I have never seen a small company that did not want to become a big company. I have never seen a small company that didn't look with envy on the research and marketing budgets of larger competitors or on the size and reach of their sales forces...

Big matters. Size can be leveraged. Breadth and depth allow for greater investment, greater risk taking, and longer patience for future payoff.

It isn't a question of whether elephants can prevail over ants. It's a question of whether a particular elephant can dance. If it can, the ants must leave the floor."

The turn-around that IBM experienced was nothing short of extreme, both in it's scope and the amount of time in which it was achieved. It took just 4 years for IBM to come back from the brink of bankruptcy in 1993 to surpassing its previous highest stock price. Since then, the company has grown to many times its previous size, and is now the world's largest and profitable IT company with revenue exceeding $100B, the majority of which is comprised of new lines of business.

Business agility is a skill, not a given because you are small.

That being said, the David v. Goliath story that gets spun a thousand times over in the tech industry (guilty here, too) is a load of crap. Whether it's the new wave of web 2.0 companies, independent game developers, or boutique PC shops, you have got to be a true innovator with lots of smart people and a little bit of luck to succeed. The market is not forgiving because you are battling bigger companies. Customers do not make purchasing decisions based on your size (and if they do, they'll go with the bigger company not the smaller). And while smaller companies have the potential ability to do very specific things better or more fully service very niche groups, don't kid yourself that someone with $$ couldn't commoditize what you do (if they aren't already!).

The only way to keep from having to leave that dance floor is to stay ahead of that elephant's feet!

Continue reading >>>

NVIDIA, Run, Don't Walk, From x86

Even though NVIDIA has been trumpeting for a year now that the era of the GPU is upon us, the fact of the matter is that everything you'd ever call a computer or a computer-like device needs a CPU. And they don't have one while both of their competitors do.

Problem? NVIDIA says no, of course. But it is. It's a big problem, and that's played out pretty hugely over the past year.

No matter what NVIDIA does while they are dependent on someone's CPU platform, they're at a disadvantage. They don't have the muscle, financial or otherwise, to grow the GPU market faster or bigger than the CPU market, and in the mean time they are going to have to continue to deal with 1) legal woes over even being able to compete at all (as in chipsets), or 2) with alleged "anti-competitive" practices stunting their ability to sell product that is legal (as in ION).

Further, the high-end gaming market that used to be NVIDIA's bread and butter is all but gone. Gone are the days of the $1500 worth of GPU's in system, a victim of the console-centrism of the games industry, the depressed economy, and AMD's onslaught in mid-range of the market. In fact, the discrete GPU business as a whole isn't what it used to be for NVIDIA, so, regardless of what NVIDIA has to say about GPU vs. CPU, the fact is that they they don't own the GPU market anyway.

There is no light at the end of the tunnel here, so I say an exit to more promising markets is in order. If NVIDIA isn't going to get into bed with AMD, they need to get to dive in headfirst-bet-the-company-style with ARM. x86, although a massive market, is not the only game in town, and the only way to truly fight Intel is to get out from under their shadow completely. Fight x86 from the bottom up!

I've been talking about the "digital divergence" for a while. In a nutshell, it means that computers will be freaking everywhere, and you'll probably use those "other" computers more than your main PC. They are in your TV, in your printer, your phone, your Zune, and probably will be wired into your eyeball before long. The market is already huge and is continuing to grow at breakneck spead now that smartphones/MID's/netbooks are taking over and Windows is irrelevant. And ARM is the driver for this stuff because it's so tiny and cheap and power efficient. It's a completely different animal from x86, and NVIDIA is the type of company that can make ARM a real contender with some real graphics technology. That's the opportunity, the door is open.

There is this fantastic book that, if you haven't read, you should: "Good to Great", by Jim Collins. Two of the chapters are focused around "Disciplined Thought" in the mind of the people in the driver seat of companies that aspire to be "great". It's defined as "confronting the brutal facts" about your business' position or situation, and having the ability to find a focus that is based around the simple, targeted thing that you can become the best in the world at.

For NVIDIA, that thing does not include x86.

Continue reading >>>

The Inevitable Bleak Future of PhysX

When Ageia hit the scene in 2002, they were a hardware company, plain and simple. The gem of their IP was absolutely their physics engine, which must have been written from the beginning to specifically shine when offloaded onto parallel processors. Their product was a physics engine, but their business model was that of an independent hardware vendor (reminds me of Bigfoot Networks a little). The rest is history, a vicious catch 22 that was insurmountable by such a little fish: Game developers didn't adopt PhysX because there was no install base, and Ageia couldn't sell their hardware because no games actually used it.

So, when NVIDIA stepped in last year, that seemed like a real good answer to the problem. But what actually happened was PhysX went from being proprietary technology with a miniscule market (that is, PC enthusiasts willing to shell out the cash for a PhysX card), to being a proprietary technology with a much bigger market -- but still too small to achieve critical mass and become a standard. That's right, 100 million CUDA-ready GPU's isn't enough, and that's because physics tech is now tied up in GPGPU and parallel computing trends, and all of that is industry-transforming stuff. Parallel computing and GPGPU is an emerging platform, and NVIDIA isn't a company in the position to control it. Ageia was too small to own the hardware-accelerated-physics market, and NVIDIA is too small to own the parallel computing platform.

Newton couldn't control physics -- neither can NVIDIA.
Platform technologies in the PC industry typically can't be proprietary when we're talking about interconnects. And really, they shouldn't be, because these types of technologies are what enable different segments of the industry to work with each other. Take PCI and USB, for example, the development of which was largely done by Intel (with industry support), and then made available royalty-free. That was the plan from day 1, because with standards in place, the entire industry was able to focus on creating products (cameras, printers, sound cards, etc etc) that consumers everywhere could easily use, instead of being focused on if consumers are going to be able to use them at all. The goal wasn't to control the interconnect, it was to give it away so the entire market would benefit and grow.

And that's really what we're talking about here with PhysX running exclusively on NVIDIA's CUDA. For a software developer that's building any product, including games, based on GPU acceleration through CUDA, they are automatically limiting their potential market. Even if NVIDIA gives all of the tools and tech to them to create their product, the whole deal is soured by their dependence on NVIDIA. Who wants their market to be smaller than it potentially should be? Why cut out ATI and Intel customers as potential buyers? What if NVIDIA loses market share? Software developers would absolutely rather be programming products for use in all parallel processing environments, be they NVIDIA, AMD, or Intel, and GPUs or multi-core CPUs.

OpenCL will swallow up CUDA software
The market has been pretty clear on what it's looking for, if you look at adoption rates. NVIDIA has generated plenty of press around its tech, but if you look at what's actually out there, there's only one title that really passes for a decent gaming title: Mirror's Edge. Anand Lal Shimpi just recently wrote a pretty comprehensive review of available CUDA/PhysX software, and the verdict was a big "meh".

But even if there was better stuff available, OpenCL (or possibly DX11 Compute) is the elephant in the corner. OpenCL is what software developers want, that's a fact, because it's hardware agnostic. C for CUDA holds no benefits over OpenCL, and since OpenCL can plug right into CUDA, isn't it pretty irrelevant then?* When OpenCL materializes, developers are going to be porting their CUDA software to it as fast as they can.

Where does that leave PhysX?
PhysX is like collatoral damage in this whole drama. NVIDIA is using PhysX as a pawn in their war with AMD by trying to use it to grow their own market share only through keeping it an exclusive technology. I think that's stunting the enormous future potential of the product itself, though, considering the market forces at work. NVIDIA should instead be pushing OpenCL like crazy, which will inevitably become a standard anyway, and grow the entire market (with their piece of the pie along with it). They'll never control the "parallel computing platform", so why waste PhysX trying?

NVIDIA should have taken a look at Intel's playbook on this one. Intel bought up Havok back in 2007, but left it to function as a viable business. It runs on all hardware, and was even spotted running on AMD GPU's via OpenCL. Havok is well-paid by developers for their engine and they undoubtedly have an exciting future ahead of them continuing to innovate in their space.

PhysX, on the other hand, is still free. NVIDIA gives it away, just like Ageia did, with the expectation that it will create demand for its products and generate profit that way. But what happens down the line when that strategy continues to fail? Will NVIDIA start asking developers for payment then? How do you think that bait and switch is going to go down?

NVIDIA should have turned PhysX into the de facto physics engine for the parallel computing revolution that's fully underway -- but as an autonomous, profitable, and viable business. I think physics is going to be a huge market force in the near future, and coupled with NVIDIA's expertise and interest in GPGPU and their relationships in the industry, PhysX could have been a cornerstone technology and key part of a massive industry paradigm shift. Instead, it's going end up a footnote, right next to CUDA.

* I think it's fair to say, though, that CUDA is irrelevant when OpenCL matures. NVIDIA has had some very cool design wins with CUDA, and there are early adopters in the developer community who were waiting for this technology and are absolutely embracing it now that it's here. So, looking at this from NVIDIA's perspective, CUDA can be seen as their way of pioneering the GPGPU space while OpenCL is being worked out. After all, they are part of the OpenCL consortium and are working in some capacity to bring that tech to the market. So, in the mean time and since they are so far ahead of anyone else in this space, why not put CUDA out on the market and use it as a competitive advantage while they can?

Continue reading >>>

No Place for Windows in the Living Room Anymore

Digital convergence is a myth, not gonna happen (correct me if I'm wrong that this isn't even being debated anymore). The error in the first place was in the interpretation of Moore's Law: As engineers were able to pack more and more power/functionality onto a piece of silicon, the sea change didn't happen at the high-end where one super-chip would have the capability to do everything. Instead, it happened at the low-end, where even ARM-based processors can now do incredible things. So it's almost like we've got the complete opposite, a digital divergence, where devices everywhere can do what only our PC's could in the past. That mentality is being translated directly into consumer electronics that exist in the living room.

That's got big implications for the Home Theatre PC as we know it. Ever since video cards shipped with composite outputs, and ATI's incredibly forward thinking All-In-Wonder line allowed TV control, capture, and time-shift on the desktop, the vision of that all-powerful PC at the TV has captured the hearts of enthusiasts everywhere. But the functions of what the traditional HTPC was doing isn't the possession of Windows anymore. In fact, anything that can run a browser at this point can fulfill the functions of a HTPC, and what people want are consumer electronic appliance-type devices that work like the rest of the stuff that plugs into their TV, not a PC. They always did. Except, now, these CE devices can actually do what we always wanted them to.

Who even watches "TV" anymore?
I haven't had cable since I was in college, and I honestly don't miss it. Much of my generation has chosen to substitute video games for entertainment instead of TV, but when we do want to watch prime-time programming, there are "other ways" to go about it. Those ways that are legal include major network websites, Hulu, Amazon, YouTube, Netflix on-demand, and iTunes -- and this is not a comprehensive list by any means. Between all of this, renting movies, and playing games, I have more than enough to keep myself occupied without spending $50-$100+ a month for cable.

So the question is really just how to get this stuff onto your TV, and honestly that's the easy part. Like I said, anything that runs a browser (which is everything) can access the content, and the vast majority of modern computers (desktops, laptops, netbooks, Macs, whatever) have HDTV support out of the box. But the devices that people actually want are super simple to use and cheap*. You can get Netflix and Amazon Video on the X-Box, integrated into some Blu-Ray players, and even on this neat little $99 box. You can get all that, plus Blockbuster and Youtube on TIVO's newest boxes. And how about AppleTV? Everything on iTunes. God help Microsoft when Apple opens up the appstore to AppleTV and people are playing games using their iPhones as controllers.

Sharing Media
One of the main things the Windows "Media Center" was supposed to accomplish was giving you a way to organize, find, and view your local content. But who is really doing that at their TV? Sharing pictures and videos is one of the main features of social networking sites, because people want to use the internet to share their content with family and friends instantly, where ever they are. You send content right into their hand on their smartphone, you don't make them come and sit next to you on your couch. And music? Having a 300GB collection of pirated MP3's is so 2001. Last.fm, iTunes, and the Zune are the obvious future. $15 / month for unlimited access to Microsoft's Zune Marketplace? Yes plz -- no dedicated NAS required.

Gaming on a Windows PC? Not if Microsoft can help it.
The HTPC was also supposed to be a way to play your PC games on the big screen. If you can get past the obvious problems with this, like the fact that the keyboard/mouse interface doesn't work unless you're sitting at a desk, you're going to come up against the fact that gaming in the living room is what consoles are for. The games were made for that environment and the prices are more in line with a CE product (after all, how many relatively expensive gaming PC's do you want to maintain in your house?). Microsoft, ironically, doesn't really care for you to game on Windows at all, let alone in your living room.

And we're just getting started with all of this. Windows as a whole, as an ecosystem, is under fierce attack, and I'm not sure Microsoft is really focused on trying to fight that battle in the living room. One thing is clear: if the traditional HTPC is going to have any place in the future outside of enthusiast circles, Microsoft would need to be making some huge moves right this minute. I don't see it.

* This is the same conversation we have around console vs. PC gaming. Consoles make sense for the mass market because those are gamers that don't give a lick about hardware and just want to play games. The console is to the gaming PC as CE devices at the TV are to the HTPC.

Continue reading >>>

Future Games Will Battle Over Physics, not Graphics

The development of game engines that aim to achieve higher and higher graphical fidelity has largely reached a plateau.  People are still gawking at the Crysis and COD4 engines, released back in 2007, because there is still nothing that has been released that looks better.  It's actually kind of hard to figure out how that would be possible, anyway, when it's basically impossible to distinguish between the game and a photograph.  And maybe that's the rub for the developers, that if they spend however many millions of dollars to try and make a better looking engine than Crysis, is that really going to sell more games and give them a tangible ROI when there's already an engine sitting there that looks so good?  Probably not.

But that doesn't mean that the quest for higher realism in games is over -- far from it.  The battleground, however, is shifting toward a focus on physics and AI as developers learn to turn picture perfect virtual environments into something that acts like the real world.  Take, for example, COD:WAW's addition of flamethrowers by Treyarch into the COD4 engine.  Or Far Cry 2, which I recently blogged about being at the absolute cutting edge of this trend: interactive flora and fauna, environmental destructibility, realistic environmental fire behavior, integration of real time and weather.   This is the future of high-end software.

There's a raging war, though, around this technology that will define the future of the market, and it's aggressively fed by the dealings of Intel, AMD, and NVIDIA.  These guys all use game developers like pieces on a chess board, and big design-in wins are expensively bought and widely celebrated in the media. They've all got their angle -- NVIDIA has PhysX running on their GPU, Intel has Havok running on their CPU, and AMD is porting whatever they can to their own hardware.

But this segmentation won't last long and the trend is very rapidly driving toward universal compatibility on the hardware side.  The gaming industry as a whole favors as much standardization as possible so that developers can concentrate on making great games, not playing with proprietary hardware or technologies.  Therefore, it's in the best interest of everyone involved to just get their physics engines out to the market and in use as widely as possible. Standards like OpenCL, fueled by the GPGPU revolution, are unifying the market and accomplishing just that.

But why is this so important to these hardware behemoths?  Because their technology is so far ahead of the software that game developers are putting out at this point, that no one needs high-end hardware to play games! Over the past year, NVIDIA and AMD have had to start seriously looking elsewhere for their paycheck, eyeing mobile and embedded markets hungrily for what has become a very lucrative and growing source of revenue.  They've been putting significant resources into growing those markets, but the high-end is where the margins are, and they don't want to lose that.  Physics technology has the potential to create new demand for high-end hardware when consumers start really looking for hardware acceleration for those functions in games.  When that happens, high-end hardware might actually become relevant for the mainstream again -- and everyone in the business would love to see that happen.

But the onus right now is on the game developers.  Barely over a year ago, before NVIDIA bought Ageia and its PhysX engine, no one was really paying attention to physics functionality.  But now, the technology is there, it's maturing fast, and it's being increasingly adopted by the industry, which means that those who are smart and creative enough to leverage it will have an edge.  When you've got massive budgets and many years of development time to produce a product, you've got to do everything you can to differentiate yourself.

So, when a company like Ubisoft integrates these advanced features into Far Cry 2 in a way that really changes the way you play the game, everyone loves it.  When you can get even a casual family game, like Boom Blox, to integrate physics into its core mechanics and also create a great experience, it's going to end up being one of the best third party Wii games out there.  When you tell me that Red Faction: Guerilla is going to allow destructibility of nearly every part of the environment, I'm going to be buying your game at launch (and it better deliver).

Graphics are old news, Crysis and the Wii have shown us.  Physics is the new graphics.

Continue reading >>>

A Chassis Ahead of its Time: GTR Tech GT3

The GT3 by GTR Tech. I heard about this product at the beginning of 2008, I have loved it from the first second I saw it, and I've been using it as my main tower for 6 months now. Basically, it's a chassis that is volumetrically smaller than the vast majority of MicroATX enclosures out there, yet it can take a full ATX board and a full-sized double-slot video card. Just let that wash over you for a second. This is tech art. Feast.

GT3 is the brainchild of Sean Hall, the owner of GTR Tech. He's got a deep background as an engineer, and if you get him on the phone, he will talk to you about the GT3 all day long and still be just as excited about it as he undoubtedly was over 4 years ago when he first started his company. We were talking a few days ago, and he said that he is renewing his push to widen the market for this product, he's working on some updates and changes to the design, and he's even got some new products potentially in the works. Pay attention to this guy!

I'll leave it to real hardware websites to handle legitimate reviews of stuff like this, but I want to highlight the significance here. Dwight Silverman at the Houston Chronicle reviewed the HP Firebird this week: "What’s interesting about these specs is that, on paper, some of them aren’t that good. The hard drives, for example, are the slower, 5,400-RPM models you’d find in a notebook. Most performance-based desktops use 7,200-RPM drives; some even use 10,000 RPM drives. The memory is also not state-of-the-art, since it’s DDR2 instead of the newer, zippier DDR3. But put the whole package together, and it works well."

Works "well"? Not good "on paper"? Firebird is a essentially MXM on the desktop done wrong. Anyone else think "good enough" isn't the type of thing you want to pay a premium for? I mean, I'm all about leveraging the fact that hardware tech right now is significantly more advanced than what games can utilize, so you can play with lower end hardware and be happy. But why pay more for laptop components to do it? GT3 is smaller, or at least the same size as, the Firebird, you can upgrade it yourself, it's more powerful for less money and it does NOT have an external power brick. Where's the Firebird's advantage again?

GT3 is for everyone. It's priced at $189 and uses industry standard components, so it can be the platform for some incredible $1K range systems that can run with the big boys. I've got an e8600/4GB/ATI 4850 in mine. Sean says he's got 4870's running in his lab, and he's done higher-end builds for customers, too. Like I said, keep an eye on him, and I hope he makes a killing because it's about time this thing got out there.

Here's to one guy taking HP's $3B R&D budget to school!

Continue reading >>>

All Your PC Boutiques Are Belong to China

Well, it's finally happened. Alienware has gone the way of the Sood, and is integrating operations into Dell, "putting two teams together into one unbeatable team."  A lot of people will get really worked up about this and pontificate that Alienware is doomed and is losing their soul (or their "DNA", for the very savvy), etc etc. But the fact is that this is the industry now, and there's no stopping it.  If Alienware had a soul, it was gone a long time ago.  Area 51 was a hoax!

Yet the funny thing is, it doesn't even matter.  No one cares who builds their computers, as long as they get the product they expected to get.  PC's, even gaming PC's, are commodity products, and no PC company of any size is doing the actual system assembly.  Not HP, not Apple, and not even Dell.  It wasn't always this way, but it is now.  So, Alienware builds aren't even going to end up in a Dell factory, because Dell is selling all of their factories!

The real scandal here, though, is that both Nelson Gonzales and Rahul Sood told everyone at the time of their acquisition that they'd stay the same sexy boutique companies, except with all of the benefits of the big OEM.  Gonzales said "this acquisition [will] only succeed if Alienware is structured as a separate division," and that "you're not going to see a lot of changes after this announcement."  Rahul Sood said "This is a deal about innovating our product line, not our supply chain" and "Voodoo will remain in Calgary" and other funny stuff about products never going into retail.  The few customers who did think there was alien technology and voodoo magic inside those boxes must be really upset.

But you know what, the reality is that the Alienware business model, copied many times over, is a freakin dinosaur.  That's the elephant in the corner for everyone who is still trying to make money spinning retail components under a slick-looking brand.  This just doesn't fly anymore, because branding itself is largely impotent in today's markets unless your company is a force of nature.  The boutique market that was started when hardware enthusiasts were competing against bland Dell boxes is over, and every piece of the puzzle of building the high-end boxes has been commodized into off-the-shelf sku's.  Alienware and Voodoo are nothing more than stamps in the mold.

For that reason, I'm going to make the prediction that the top execs from both Alienware and Voodoo don't last very much longer at their respective big-box companies.  Whether it's the restless entrepreneur inside of them, or the fact that they just can't consistently create products that are actually viable for $100B companies, these guys will move on at the first sign of greener pastures.

However, there's still a market, albeit small, for companies that are service-focused in the boutique segment.  As long as people have to deal with screwups like Acer's Predator catching on fire, there will be people wandering the internet looking for the businesses that will hopefully provide them with a higher-touch, more personal experience.  If the remaining niche boutiques can concentrate their efforts on service versus trying to build a brand or pimping the hardware, they will be fine.

Continue reading >>>

Bigfoot Networks' New KillerNIC Needs Bigger Vision

The announcement yesterday was that the guys over at Bigfoot have launched their 2nd gen KillerNIC, the Xeno. This is exciting because these are the guys that made a network card hardcore, complete with oversized knife-shaped heatsink*. They've spent the past 3 years trying to convince everyone that there is a real war to be waged against online gaming lag, and they've finally released their newest product.

I'm a little disappointed, though. The new card, the Xeno, has limited differences in relation to the original product. A smaller physical package and a PCI-Express interface are great, but those things don't allow it to minimize lag better or run better apps (since the KillerNIC is actually a Linux-computer-in-a-PCI-slot, it can run custom software that Bigfoot calls fnApps).

So, where is this whole thing going? It seems like when Bigfoot launched their original product, they showed their entire hand right off the bat. Now, they're in a spot where this brand new product launch isn't going to result in a surge of sales because they've got nothing to sell to their existing customer base and the value proposition to new customers hasn't really changed. It's like they made the decision to market their product into the quagmire of the hardware industry, but they forgot to include the piece of the puzzle that keeps companies in this industry afloat with volume: the ability to sell new/upgraded products to their existing customer base. Add to that the fact that they are selling an expensive product into a niche. Yikes!

It's worth mentioning that the past year has brought Bigfoot a new CEO, $13M in venture funding and a third-party deal with EVGA, and they've referenced the potential for non-gaming applications of their tech in the future. This certainly alludes to a grander vision than what the KillerNIC has been for the past 3 years. But maybe the KillerNIC never should have been a $250 piece of network gear. Maybe it should have been a $50-$75 computer-on-a-card that customers could BUY custom software for. That would have gotten the volume way up, broadened the customer base, and given an impetus to write innovative, super-high-quality fnApps that people would want to pay for in droves. Heck, maybe you open up the ability for third parties to write software, too. Then, you're not a hardware company, you're a platform company, and maybe, just maybe, we'd have seen KillerNIC's integrated on high-end motherboards by now.

* No, a network adapter doesn't need a heatsink, but then it wouldn't be nearly as cool.

Continue reading >>>

Far Cry 2:The Next Step in Ultra-Realistic Gaming

Far Cry 2 had me enthralled for quite a while, and I love that the conversation about this game is continuing across the community so long after its release.  I wrote about my exploits in the jungles of Africa, and about how the incredible engine and open-world environment were like nothing I've ever seen.  And that's really the clincher for me, plain and simple.  Far Cry 2's storyline is mediocre at best, and I agree with everyone that didn't give a lick about the characters, moral choices, or general narrative -- but I still think this is one of the best FPS games I've ever played.  

I called COD4 the perfect FPS last year, so what's really interesting is that the game was built fundamentally different than Far Cry -- it's based on scripted action and cinematic storytelling.  Far Cry 2 has almost none of this, and what parts of it are scripted (the intro sequence, and the areas where you accept missions) seem clunky and poorly done, completely disconnected from the rest of the game.  

Which is better?

If I had to pick either one of those titles right now, I'd pick COD4 as a clear winner for a landmark gaming experience that was culturally relevant and emotionally gripping.  Infinity Ward knows how to create that type of game, over and over again, and is going to continue raking in the dough doing it.  But, moving forward, I want more games like Far Cry 2!  Far Cry 2 has the start of an engine that is strong enough to take the gameplay experience that comes from scripted action and replace it completely with open-world, native functionality.  COD4 is the way of the past - Far Cry 2 is the future.

The Ubisoft guys are onto something. I think they've got the makings of what will be the foundations of the ultra-realistic, virtual reality games of the future, right here in Far Cry 2.  The elements are all there: complete freedom of movement around the world, photo-realistic graphics, interactive flora and fauna, destructibility (including realistic fire), integration of time and weather, and the polished feeling of a "body" that you inhabit (your character "heals" by bandaging himself, he extends his arm and you see the full motion of opening doors, he bends down and his arm picks up weapons, he becomes feint if you push him to sprint too long in one stretch, you watch as NPC's pat you down to check for hidden weapons).  I've never seen all of this done so well in one game before.  For all of Far Cry's screwups, there is alot of innovation!  When the day comes that the first truly virtual reality engine is released, it's going to be developed by these guys on a derivative of this engine.

Just a few weeks ago, I blogged about what I've seen as the recession of "graphical achievement" in games.  Certainly Far Cry 2 is one of the exceptions, and that's one of the reasons I love it so much, but it's also a great illustration as to what the future of hardware for games is going to concentrate on.  Some of the commentors on that post said that current graphics are good enough for them, and that they didn't know how it could get any better.  I say look at Far Cry 2, and let's talk about the horsepower that can be leveraged to scale this engine, not just for pure graphics, but for a more visceral environment (AI, physics).

I see the next big step forward as a closer integration of the "story" within the gameworld.  That's really where a lot of the problems are stemming from that people are complaining about (making choices, interacting with NPC's, endless respawning guards).  Make The Jackal a persistent "person" in that actually exists at all times that you have to "find", instead of just a piece of the narrative.  Make the neverending respawning guards be reinforcements that are dispatched from an actual location in the game world and have to drive to their posts.  In the scheme of things, these aren't big problems, and I bet with more development time available to them, the Ubisoft team would have had it done in FC2.  They've been very open about their process in many interviews in the media, and those guys knew where they were cutting corners to hit deadlines.

I just want to throw out there, too, that over a year ago there was a great conversation that was being had around if we can tell effective stories or have meaningful game experiences with guns.  I look at Far Cry 2 and Mirror's Edge, and see that we're watching the evolution of the FPS genre right now to a form of gameplay that doesn't necessarily require the "shooter", and places more of the emphasis on "first person".  That excites me.

And that's why Far Cry 2 was my game of the year for 2008.  These guys are my heroes right now, and I'll definitely be watching closely for the next iteration.

Continue reading >>>

Mega-Irony: Microsoft Digs Up Apple's 25 Year Old Mistakes

"Those who cannot remember the past, are condemned to repeat it."

The X360 is a cancer.  No, not because it has had a negative effect on PC gaming and I'm a PC fanboy, although all of that is true.  It's because it's a product and a business that leeches revenue, mindshare, and resources away from Microsoft's core businesses.  The XBox was supposed to be Microsoft's key into the living room -- a way for them to capitalize on the "digital convergence" that would occur there.  But it ends up that XBox business unit is really just a fortress in the living room that even the greater Microsoft will never be able to add further value to, and digital convergence may never actually occur... So it's a big ole rotten egg that has stolen from Windows its only remaining "killer app": GAMING.

In the end, the X-Box is going to kill Windows (or drive it to go open source?!) for that very reason.

Bill let this wolf into his flock against his better judgement, if Dean Takahashi's book "Opening the X-Box" is to believed.  According to it, the team who developed the product pulled a bait-and-switch on him when they got him excited about a lower-end version of a PC that could play games, but slowly morphed it into an appliance that wasn't Windows-compatible at all.  At that point, it seemed like a complicated situation that he decided to defer to his lieutenants. 

But I know Bill is kicking himself now, because he wrote the book on how to build an industry around the PC.  In fact, he spelled it out in detail to his friends over at Apple all the way back in June 1985.  John Sculley was at that point firmly at the helm of the company, and Bill, as the biggest developer of software for Apple at that time, wrote him a memo outlining what he thought should be Apple's go-forward plan: licensing the Mac's technology.

Apple's stated position in personal computers is innovative technology leader. This position implies that Apple must create a standard on new, advanced technology.

Apple must make Macintosh a standard. But no personal computer company, not even IBM, can create a standard without independent support. Even though Apple realized this, they have not been able to gain the independent support required to be perceived as a standard.

The significant investment (especially independent support) in a "standard personal computer" results in an incredible momentum for its architecture. Specifically, the IBM PC architecture continues to receive huge investment and gains additional momentum. (Though clearly the independent investment in the Apple II, and the resulting momentum, is another great example.) The investment in the IBM architecture includes development of differentiated compatibles, software and peripherals; user and sales channel education; and most importantly, attitudes and perceptions that are not easily changed.

Any deficiencies in the IBM architecture are quickly eliminated by independent support. Hardware deficiencies are remedied in two ways: expansion cards made possible because of access to the bus (e.g. the high resolution Hercules graphics card for monochrome monitors), [and the] manufacture of differentiated compatibles (e.g. the Compaq portable, or the faster DeskPro).

The closed architecture prevents similar independent investment in the Macintosh. The IBM architecture, when compared to the Macintosh, probably has more than 100 times the engineering resources applied to it when investment of compatible manufacturers is included. The ratio becomes even greater when the manufacturers of expansion cards are included.

The companies that license Mac technology would add credibility to the Macintosh architecture.  These companies would broaden the available product offerings through their "Mac-compatible" product lines. 

They would each innovate and add features to the basic system (various memory configurations, video display and keyboard alternatives, etc.) Apple would leverage the key partners' abilities to produce a wide variety of peripherals, much faster than Apple could develop the peripherals themselves. 

Customers would see competition and would have real price/performance choices. Apple will benefit from the distribution channels of these companies.

The perception of a significantly increased potential installed base will bring the independent hardware, software, and marketing support that the Macintosh needs.

Apple will gain significant, additional marketing support. Everytime a Mac compatible manufacturer advertises, it is an advertisement for the Apple architecture.

Can you fathom this?  It's the business plan for what the XBox should have been!  Licensing, a rich ecosystem of hardware, partners, partners, and more partners!  All the rationale for not building a closed platform is right there.  He wrote this to a company (Apple) that thought that having a lock on the most advanced and best designed product was all one needed to succeed. But we know better, judging from Apple's history into the late 90's as a story of innovation and invention with botched execution over and over again.  And Bill knew better, too.

Ideas, products are nothing without the right execution. Microsoft should never have strayed from their Windows-centric business model.

Doh!

Continue reading >>>

NVIDIA: Shift In Relevance Away From Games

I'm playing in beautiful 1920x1200 res on a 24" LCD, full detail on Dead Space and Far Cry 2, on roughly $800 worth of hardware (e8600, GF260, 4GB). I don't need to spend any more than $250 each on CPU or GPU, which means I certainly don't need SLI/Crossfire or anything like that. And the problem is that that's not going to change anytime soon. After all, there will never be another Crysis or Doom 3, where these brilliant technical developers are pushing the bar on gaming technology. These guys don't care anymore about that stuff, because no one pays them for it. They end up building crappy games with the tech that they develop and no one buys it, so they go make console games. It's sick how many times that happens.

That's what the market-at-large wants, though, right? The vast majority of consumers don't want to spend a lot of money on their PC, even to run these "technically" beautiful games (I've got people on twitter asking me if ION is a good gaming platform, for goodness sake), so game developers are just listening to their customers and writing software that their customers want, because they want to make money just like everyone else. Who's going to fault them for that?

But that means that at least 30%+ of NVIDIA's business, the hardcore PC gamer, has no reason to spend any money, and NVIDIA (and anyone else that makes high-end PC products, like Intel and AMD) is in big trouble. The current hardware survey on Steam shows only 1% of gamers have more than one GPU, and the most popular GPU by a landslide is still 8800 series. We need games that will take advantage of as much power as is available in hardware, so we can have some reason to sell high-end GPU's. Recession or not, the games are the problem -- our "killer app" is gone!

No wonder Jen-Hsun is up there with Charlie Rose waving ION around! He goes, "This is the Atom processor. It's my favorite processor in the world." lol! NVIDIA is running full speed away from the gaming market and into the arms of GPGPU and IGP's! And as Jen-Hsun talks about what the paradigm shift of relevance for his company is, every god-fearing, self-respecting PC-head looks into the dim, brooding future and sees the dark age of PC gaming looming over him, with cackling, half-naked imps holding XBOX controllers dancing around him.

There's a lot of people who do want to pay for this stuff, but there is just no reason to right now. So, maybe NVIDIA needs to buy some developers and start making some games (I promise I'll buy them, even if they suck), because no one else in the gaming world gives a crap whether they live or die right now. They've all got their own problems... and NVIDIA knows it.

Continue reading >>>

Microsoft's Blunder of the Decade : What X-Box Should Have Been

In 2007, Bill Gates and Steve Jobs appeared for the first time together in a live joint interview at the All Things Digital conference, answering pointed questions about the history behind their companies and thoughts on the current state of the industry. At one point, the conversation turns to hardware vs. software business models, and goes like this:

Steve Jobs: Alan Kay had a great quote back in the ’70s, I think. He said, "People that love software want to build their own hardware."

Walt Mossberg: Well, Bill loves software.

Bill Gates: Oh, I can resist that. --everyone laughs--

Microsoft is a software company.

You get a real glimpse into Bill's mind in that conversation, and it's obvious his business was, and always has been, software. That is what Bill is passionate about and that's what is at the core of Microsoft's DNA. Not hardware. Hardware is simply not their business, and they became the behemoth that they are because they had the right approach for tapping into the latent talent across the industry : create a standard, based around their software, for all the other PC hardware manufacturers in the industry. This model works and works well.

Steve Jobs makes this comment later on in the same interview:

You know, because Woz and I started the company based on doing the whole banana, we weren’t so good at partnering with people... where Bill and Microsoft were really good at it because they didn’t make the whole thing in the early days and they learned how to partner with people really well.

That's why the model works so well. Microsoft is great at working with partners and that's another key element that is infused into their core. But all of that is the old Microsoft. The new Microsoft chases the around the iPods and the Playstations of the world, playing the game with their competitors' rulebooks. And that thinking is what has led them down the path of one of the biggest mistakes in the life of the company: The X-Box.

Microsoft vs. Microsoft

The Microsoft Gaming business unit lost a piece of the company's spirit when they made the decision to make an X-Box at all. The fact is this: Gaming PC's come in every range of price, shape, and ridiculousness -- but there is one, and only one, constant in all of them: Microsoft software. And Microsoft decided that instead of fostering that huge, massive customer base that they already have in place (basically every PC in the world that runs Windows) in the proven way that they've been successful for 30 years, they'd instead be a competitor... to themselves.

That is the blunder that has cost Microsoft not only billions in losses, but billions in lost revenue they could have had if they would have embarked on a completely different business trajectory. Microsoft never had to develop their own hardware -- after all, they downright suck at it. They could have spent that time and money working with partners to accomplish the same end (provide a gaming platform), with none of the risk, and all of the reward.

Instead, Microsoft has chosen to make their support of gaming on the PC a joke. After Microsoft's announcement of their poor numbers for last quarter, they started making cuts. And the first to go? The head of the Games for Windows business unit, and Aces, the developer behind the 25+ year old Flight Simulator franchise, a cornerstone of the culture and history of PC gaming. In fact, Aces, and Ensemble (Age of Empires developer) before them, are the last in a string of game developer cuts since 2006 that leaves Microsoft with no developers left for the PC at all. Fathom that.

As a result, Microsoft now faces real threats from companies who plan to make Windows irrelevant. Whether it's Linux on netbooks, Google in the cloud, or OSX on the Mac, Microsoft is losing Windows mindshare with the masses -- those same people who are buying X-Box's, which Microsoft makes no money on (instead of the 22 million copies of Windows they could have sold). Now, when those people are looking for a replacement for their current Windows PC, Windows won't be on their list of requirements when they buy. After all, they game on their X-Box, now they just need something to run a browser on. Earnings last quarter on client sales revenue: down 8%. X-Box and PC games revenue: down 22%.

If you want an easy way out, blame it on the economy -- but please disregard Apple's amazing last quarter growth, the 5 bajillion netbooks that sold last year, and the fact that gaming industry revenue continues on an explosive trajectory.

WHYWHYWHYWHY

Someone at Microsoft talked someone else at Microsoft into the console business model that says: 1) Need cheap hardware to grab massive and widespread market penetration, 2) To get it cheap enough, have to take a loss on the hardware, 3) We'll make it all back on the software.

That guy probably doesn't have a job anymore.

But a new guy in his place is thinking the multi-billion dollar mistakes have all been made, so it's time to start making money. That remains to be seen.

The problem is that Microsoft didn't think that third-parties would fit into this. For this to work, someone has to make some serious concessions on making money on hardware to get the price of entry down. I suppose Microsoft thought they would be the only one interested in doing such a thing (or it was just plain, stupid greed, which is more probable). No!! Partners!! Please don't tell me that someone can't come up with a revenue sharing model where even games sold in retail could be tracked (digitally distributed software is obviously much easier). With all those $billions$ they'd have to spare from not wasting it all on the X-Box hardware, Microsoft could certainly crack that nut. $Billions$ could certainly have worked wonders on the piracy problem. $Billions$ applied to the right problem goes a long way.

Partners!!

What the X-Box should have been

It didn't have to be this way, though. The X-Box should have been every Windows PC on the planet. It should have been a digital distribution platform to rival Steam. It should have been a Windows Gaming OS sku that looks exactly like what you see when you turn on an X-Box. It should have been been Microsoft leveraging the long-tail of indie PC development and bringing that talent into the mainstream, along with their big AAA exclusives like Halo and Gears. It should have been hardware manufacturers competing over console system designs -- I should be able to buy an HP or Dell "X-Box" right now, and I should be able to leverage the fact that those manufacturers will give me options for Blu-Ray or Wifi and whatever else they would need to do to compete with each other and Sony. The consumer wins. Microsoft wins. Partners win.

Everyone lost

Microsoft opened their own can of worms on this one, and frankly, they can have it. But, every partner who stood to benefit from Microsoft not screwing everything up should be pissed. Instead of building a rich ecosystem like we've got on the PC, we've got a console vs. PC war that nobody benefits from except for the forum fanboys who thrive on drama. Now, I can't play Halo or Fable, and Blizzard can't sell WOW to 22 million X-Box gamers. Now, X-Box gamers get crappy networking and "matchmaking" for multiplayer (console gamers don't even know what they're missing), and I get the joke that is Games for Windows Live. Now, X-Box gamers can't even use a browser or access the huge libraries of classic games from GOG.com or Steam, and I can't play XBLA games.

Actually, everyone didn't lose. Nintendo is pretty happy.

Related article: "Mega Irony: Microsoft Digs Up Apple's 25 Year Old Mistakes"

Continue reading >>>

How Many Video Cards Do You REALLY Need?

I remember how hardcore I thought it was when SLI and Crossfire were just coming out. The thought, back then, that PC's were going to be able to push the bar that much more from a hardware point of view seemed like a total breakthrough. I'm a PC hardware nut, and I can totally relate to the mentality that loves adding more and more crap into a system just for the heck of it. LOOK HOW FREAKING BIG THOSE THINGS ARE, and there are THREE of them!!11!!1! Do you fathom how much POWER I need to run this thing?!

I love Skulltrail, I love ridiculous add-on cards, I love hotrod chassis...  But that's one class of people. And as far as PC users go, even in the gaming niche, they are a minority.  It's not a secret, anyone can go right to Valve's website and look at what millions of gamers are running on right this second.

I recently read a piece that puts this very well:

The days of needing multiple graphics cards in your PC are officially over. While game engines have become more complex, and Crysis still confounds most people’s PCs, the fact is that we have reached a pixel density plateau when it comes to monitor sizes and the GPU battle grounds are being fought at 1680×1050 resolutions or lower. And the simple fact of the matter is that at that resolution nearly every game on the planet can be enjoyed by the mainstream gamer utilizing a single graphics card. The hardware side of the gaming industry has gotten that good.

This isn't about NVIDIA or AMD, or consumer demand for the latest and greatest, or whatever. It's about reality.  This has huge relevance for the platform wars and the perception that the PC is battling against day in and day out.  People need to understand that you don't need to deal with the finicky hardware issues, less-than-perfect driver support, and high price tags that come along with running multiple graphics cards.  If you're the average gamer (95% of PC gamers are not hardware enthusiasts, this is not a secret), you can run the vast majority of GREAT games out there at settings that are going to look better than what you're seeing on any console, and you can do it not only with a single graphics card, but with a graphics card that you can afford.

Seriously.  This is a good thing. 

Continue reading >>>

CliffyB: "You're All Pirates, No Gears 2 For You"

CliffyB, a guy who doesn't have the strongest track record of thinking things through before he talks, recently had some new things to say about the PC as a gaming platform and EA's plans for Gears of War 2 on the PC.  For Cliff, the "PC is different now than it was back in the day", and he doesn't want to deal with problems that arise around designing games for various hardware configurations and sub-par integrated graphics.  And those gamers who are savvy enough to maintain machines that are able to play his games?  Well, those people are simply pirates, of course.  As far as Cliff is concerned, if you can upgrade a video card, you can hop on BitTorrent and steal his games!

To anyone with some sense and some perspective, this is flat-out nonsense.  The problem is that CliffyB's lunacy is obviously contagious within Epic, and therefore Gears of War 2 will not be coming to the PC.  At first thought, that doesn't sit all that well, considering what a great game Gears of War is and how psyched I am for the sequel... but then I remember how crappy the PC port was and it doesn't bother me all that much anymore.  I have no doubt Gears 2 would be similarly poor on the PC, so, yeah, why bother?

The point here, though, is that Cliff has got it wrong, his strategy is flawed, and therefore he's not maximizing the value of his game.  Brad Wardell of Stardock wrote an article a few months ago that shows how a real businessman in this industry should be thinking about this problem.  After outlining how two of his games, without any DRM at all, outsold many big budget, widely marketed titles, he says,

"Our games sell well for three reasons.  First, they're good games which is a pre-requisite. But there's lots of great games that don't sell well.  The other two reasons are [that] our games work on a very wide variety of hardware configurations [and they] target genres with the largest customer bases per cost to produce for."

He's talking about people like CliffyB here.  Cliff has got a good game, but he's lost the ability to attract the PC side of the market.  Maybe Epic just doesn't know that "gamers" aren't one big pool anymore and that the market has grown to a point that they actually have to engage the PC side in a different and purposeful way.   If you want to make money off of PC gamers, you've got  to solve the compatibility problems, tailor the graphics and controls to the strengths of the PC, and give exclusive content or features that creates a great experience that is unique to the platform.  I mean, it's almost a standard formula you can just follow and implement.  Just do it!

Roy Taylor from NVIDIA did an interview a while back where he said that games are trending toward starting their life on the console, then being improved for the PC.  I agree that's the way it should be, and Mass Effect was a great example of this.  Widely acclaimed as far better than the X-BOX original, Mass Effect was given a significant overhaul by a third-party, Demiurge Studios, during the port's development.  I mean, these guys should get a medal for proving that it's possible to take a game that was made for consoles and create a product that PC users will want to spend money on. They added native, intuitive control for the keyboard and mouse,  completely overhauled all the menus, and added free additional content, and because the game runs so much better on more capable PC hardware, the port was a hit.

And here's the clincher, Mass Effect uses the same Unreal Engine that Gears of War does. 

So here's the lesson for CliffyB:  You can't hate on your customers.  No self-respecting PC gamer can stand him and his consistent negativity toward everything PC.  You'd think that single-handedly creating a massive wall between yourself and millions of potential $buyers$ would be grounds enough for someone to tell the guy to get a clue already...

Continue reading >>>

Turning Screws Is So Tedious: Neo-System-Integration

Buzz buzz buzz "HP Shutting Down Voodoo!" was the headline burning across the net this week (power supplies being returned are apparently indicative of armageddon these days). As the whole story got shaken down, it turns out all the noise was most likely instigated as employees started getting layed off from Voodoo's Calgary facility, which is probably the forerunner of that location getting severely downsized or even shut down completely. What does it all mean? Phil McKinney, HP's CTO, explained that it's the start of Voodoo's "integration into the larger HP".

One could argue that when HP, a company that sells more PC's than anyone in the world and yet outsources the assembly of over half of those systems, buys an SI that builds 100 systems per month, they probably aren't looking at the manufacturing infrastructure as something that's a very valuable piece of the puzzle. HP bought a brand to compete with Dell's acquisition of Alienware -- I'm frankly surprised it's taking them so long to "integrate". I mean, after two certainly transforming years under HP, how much of that "VoodooDNA" belongs to Calgary? Moreover, how much of HP Gaming belongs to VoodooDNA? Moreover, who came up with something as ridiculous as VoodooDNA?

Many people have been pretty critical of the way that Dell handled the Alienware acquisition, basically leaving them alone to compete with their own XPS line. I think those same people might see Voodoo being relegated to the status of brand monicker as cause for thinking maybe Alienware did something right after all. I know the Alienware employees that still have jobs are seeing it that way at least.

However, Alienware's future isn't so sure, either. Earlier this month everyone was talking about the fact that Dell wants to sell all of their factories. It seems that their supply chain just isn't so unique in the last few years, and with $3B that needs to get cut out of the budget ASAP, Dell is has apparently bigger things to worry about than pesky annoyances like building computers. If Alienware gets caught up in that, maybe they won't look so different from Voodoo after all...

All of this asks the questions: How much of the boutique's spirit is in the huddling over a bench turning screws? Maybe a better question is, How much of that is what customers care about?

Continue reading >>>

The Future of Gaming is Multi-Core

Whenever we're talking about price/performance configurations for gaming, it's generally well-known that an all-around good way to handle that is to go for higher clocked dual cores over lower clocked quad cores (which would both usually be comparable in price). We do that because more games do not take advantage of multiple cores than do, so we're putting the customer's funds where they'll be used most optimally. (That's actually the premise of nVidia's "Optimized PC" campaign -- except NVIDIA wants to load up on GPU power specifically, but that's another conversation.)

All of this is changing, though. As the raw number of cores in multi-core processors continues to scale higher and higher, software developers are compelled to optimize their products to utilize the technology. Just like NVIDIA has a dedicated team of people who manage developer relations and are tasked with making sure that devs are building support for the latest GPU tech into their games, Intel has a team that is undoubtedly doing the same thing. Multi-core gaming is here now.

I started getting very interested in multi-core gaming around the launch of Skulltrail. Yes, that platform was largely a flop for the gaming space, but it was the first time that I had heard at all about games that were being optimized for up to 8 cores!  At that time I got my hands on a short list from Intel on specific titles that were able to take advantage of many cores, and on that list was Far Cry 2, a particularly exciting title that's getting very close to release.  This game is going to be in every system reviewer's benchmark suite after it comes out, and it looks like it's going to be Intel's gaming poster child: dynamic weather system with air patterns that affect foliage and clouds, super-advanced AI in enemies and animals that reacts ultra-realistically to environmental and player actions, and the Havok physics engine -- all running as separate processes on their own cores.

That's not all.  Tim Sweeney from Epic recently talked about how the Unreal Engine scales "very well from dual-core to quad-core, and actually you can see a significant performance increase" when using quad-cores.  The Unreal Engine is probably the most-licensed engine in the industry, and that means that a massive amount of current and future games are going to be seeing the same performance gains from multi-core. 

I'll continue keeping my eyes open for more info around all of this as time goes on, but the real point here is that we're getting real close to the point where general hardware spec recommendations are going to start favoring more cores for gaming.  In the mean time, paying attention to the nuances of the specific software you'll be running is the best bet -- you might already be running titles that are multi-core-friendly.

Continue reading >>>

Microsoft on the Attack: The PC is Cool Too

"I fear all we have done is to awaken a sleeping giant and fill him with a terrible resolve." - Isoroku Yamamoto, commander of Japanese forces at Pearl Harbor. (Actually the quote is probably fabricated, but who's going to argue with Ben Affleck's dreamy smile?)

Microsoft is on the move lately -- Bill is pissed.  Do you think Steve Jobs is scared? Does Steve Jobs get scared?

I jest, but the truth is that the strength of the PC industry as a whole has a lot to do with the image that Microsoft creates. When Apple is out there for years with their smear campaign, that negative consumer perception trickles down into every niche that uses Microsoft software as a component. If our customers think Windows is unsafe, slow, incompatible, or any other flavor of unpleasantry, then no matter how great our product or service is, we're at a disadvantage. That's why I think Microsoft's new surge of energy to get out there any win hearts and minds is a great development... and it's all Apple's fault.

First off is the most obvious Seinfeld commercial that we've been hearing about for months.  It made me laugh, but that's because I'm a tech nerd... I'm not sure the general public got anything from it.  The message just wasn't as clear as "I'm a Mac.", "I'm a PC.", "PC, you're fat and you suck.", "Yes, I know."  But, it's a step up from nothing at all, and I'm assuming the next few commercials will actually have a point.

Next thing is Microsoft making a "fashion statement".  That's right, Microsoft at PROJECT Las Vegas, a global tradeshow for the fashion industry.  What does it mean?  It means that Microsoft is not only aware, but is trying to push, the fact that PC's don't have to be the second-rate alternative to great Apple design.  Microsoft doesn't sell hardware, but they are investing time and money to shape the perception of the physical appearance of the PC.  This is cool.  Love it.

Another thing Microsoft is trying to do to control the perception of its product is to put Microsoft "gurus" into retail, a la Apple "Genius Bars" (as if that isn't the most egotistical thing I've ever heard of -- how do you even go near that thing without feeling wierd?).  This is a direct, high-touch way to get face-to-face with Microsoft's end-users, something the market hasn't had before.  General consumers, who have only seen Microsoft as the "tax" they've had to pay on every PC purchase they've ever made, can now be woo'ed by product experts that hopefully can not only ease their fears into making the jump to Vista, but show them WHY they'd want to do that, and maybe even answer technical questions post-sale.

Next on the list, Microsoft-branded casinos and men's clubs.  You know Ballmer's all over that.

Continue reading >>>

id's Hollenshead: "Hardware Manufacturers Are Happy About Piracy"

In a recent interview at QuakeCon, Todd Hollenshead, CEO of id Software (the developer of the first FPS games, Wolfenstein and Doom), made some pretty odd comments. When asked to clarify his comments about PC hardware manufacturers being "happy about" what they considered the "hidden benefit" of PC piracy, he answered, "Yeah, I think they are... They realise that trading content, copyrighted or not, is an expected benefit of owning a computer... Like a right to download music for free or a right to download pirated movies and games."

...Ouch.

See, the problem with stuff like this is that its such generalized, non-specific nonsense, yet so antagonistic and inflammatory. The minute you read something like that, you wonder who he's talking about, and why they would feel that way. Are "PC hardware manufacturers" someone specific, like Intel or AMD, or does he mean system integrators, or basically just the entire PC industry altogether? And why?

The "why" is the really telling part, and it cuts to the root of the PC piracy issue: Hollenshead thinks the PC industry benefits from piracy in some way, but the truth is that no one benefits. Pirates aren't buying PC's because piracy is some "killer app". These aren't people who are buying more PC's or better PC's because piracy is possible for them. It's like equating the PC industry to arms dealers profiting off of warfare....except we're talking about 12 year olds. It's just pushing the whole thing too far.

The point is that Hollenshead is pissed. Piracy is the nastiest problem facing PC gaming, and he's particularly frustrated as a pioneer of the industry as a whole. That's fine, and I support the cause whole-heartedly, but it would be far more constructive to be having discussions around how the hardware manufacturers can be helping the industry to overcome this specific problem, not making blanket statements which can't be substantiated. I think hardware manufacturers are in fact particularly interested in doing everything they can to support the platform. After all, the gaming industry is nothing without the hardware side, yet gaming is a source for massive profit for the PC industry -- it's a synergistic relationship.

I can forgive him, though. After all, I do love me my FPS's.

Continue reading >>>

You Know You're Over-Hyping Your Franchise When...

1) You're "developing" for consoles that don't exist and for which you have no information. Epic's President Mike Capps says that Unreal Engine 4 "will exclusively target the next console generation, Microsoft's successor for the Xbox 360, Sony's successor for the Playstation 3 - and if Nintendo ships a machine with similar hardware specs, then that also. PCs will follow after that."

2) You're projecting a launch date 10 years into the future. Capps says, ""We've got Unreal [Engine] 4 in production right now.. Our rough guess is 2012 [to] 2018."

3) You've got one guy working on it. Epic's Tim Sweeney says, "Basically, it is just me, but that team will be ramping up."

Aren't you excited?

Continue reading >>>

Non-Proprietary Hardware Means High Quality... Right?

It occurred to me recently that the claim that "assembly line" PC manufacturing is of inherently low-quality isn't the only fallacy in the boutique side of the industry. A lot of SI's talk about the usage of "non-proprietary" components as a sales bullet as well, and I'm coming to the realization that that really doesn't matter either. Customers want reliable products, great performance, and responsive support. None of that relies on using "off-the-shelf" parts at all, and if you don't believe me, just take one look at Steve Jobs. Apple, who's systems are the very definition of proprietary, is consistently grabbing more and more market share from the PC, and customers are paying more for their products on top of it. Figure that one out.

Consider where pushing non-proprietary-hardware-based systems came from in the first place: Big OEM's started pumping out garbage machines, poorly designed and proprietary. When the recipients of those machines found out the hard way that they couldn't get any help from anyone but the OEM because of the proprietary hardware, well, that was the end of that. Competing systems that used off-the-shelf hardware had the ability to be serviced by any computer shop in town, freeing the customer from the shackles of being locked into talking to call-centers in India. It's a simple, logical train of thought...

But the problem really isn't the proprietary nature of the hardware, is it? That's actually just a round-about way of fixing the real problem, which is poor design. If the hardware was top-notch in the first place, for argument's sake let's say it was even better than off-the-shelf hardware, and it was backed up by quality support, then no one would complain and the customer would get a better product. That's what customers want. Proprietary or non-proprietary is irrelevant.

Continue reading >>>

The Key to Craftmanship and Quality in the SI Biz

It's a pretty common in the boutique circle to bash the OEM by saying that a "hand-built" computer is better than some "assembly-line, cookie-cutter" box. It sounds really good, because it conjures up the image of a super-technician huddling over your beloved new machine, thinking to himself how awesome this very computer is going to be, because he's building it for [insert your name here], which in a single act will be the pinnacle of his life's work... No. The assumption that "hand-built" means higher quality or that it's in some way more suited to the individual customer than something that is "mass-produced" is a fallacy. Instead, I actually think it's one of the key deficiencies of the boutique model.

The fact is that every boutique that has plans for being in business long enough to think about any of these problems is thinking to themselves every day about how they can build machines cheaper. They are trying to figure out how to do it faster, what parts of the build process can be done by high-school kids on minimum wage, what they can automate, and how they can get components for better prices. OEM's are already 1000 steps ahead in this process, and it becomes simple economies of scale: the bigger you are, the cheaper every single product is to make.

Further, every customer that is looking at a boutique to build their machine is thinking about how fast they want to run away from anything that resembles the experience they had with their last PC from some big box manufacturer. They don't want a product that is prone to failure because of poor quality and they want to be pushed to some call center in India when they need support.

So what's the answer? Does having one physical person build your machine really make a higher-end product? Isn't it easier for one guy to screw up than ten or twenty people on an assembly line? You see, an assembly line can be taught an exact process down to the tiniest detail, even to where every single cable should go. Quality control is built into the process. So, the ultimate ideal is to teach the OEM infrastructure how to produce a boutique product. This is a win for the customer, because they have access to high-quality products at prices that they can afford.

Continue reading >>>

EA Selling Pre-Built "Crysis Warhead" PC's

According to a Gamasutra editor, via Destructiod, a rep at EA's Comic-Con booth spilled the beans about a pre-built PC that will be marketed as "Crysis-ready". I'm not surprised, I'm actually a little ticked that I'm not the one making that deal. This is something I've wanted to do for a while and it's exactly along the line of what's got to happen to start diffusing some of the negative press around the PC gaming industry.

That system obviously isn't geared toward the boutique buyer. But what about the guy that just wants to get a reasonably priced box that he knows will run Crysis (albeit not at insane resolutions, which this customer doesn't care about), and therefore will probably do decently well at the rest of the PC-exclusives that he's had his eye on? This customer is the gamer that doesn't give a darn about the hardware, but he is by no means not "hardcore" -- he's just a hardcore gamer.

EA's onto something here. The market for this product is out there.

Continue reading >>>

Which Comes First in Tech: Solution or Problem?

It's not just ray-tracing that's going to benefit from the "mass-core" -- Intel will make sure of that. A lot of the time public perception of the industry is that tech companies -- even monsters like Intel and nVidia -- create products and just "give" them to the market. Certainly, the advantage of the PC industry is that products are engineered to be able to be used in as wide a range of applications as possible, but that's not the end of it. Technology is nothing without a customer. The fact is that the competitive tech landscape isn't driven by a bunch of passive well-wishers, it's manhandled by aggressive corporate sales machines. That element is frequently overlooked by naysayers, and that's got to be considered when all of this speculation is flying around about the future of the industry and what will or will not happen.

Case in point: Dreamworks recent announcement that they're moving to Intel as their new provider for processors in their massive render farms. The point here is that a lot of people are trying to say that we've "got enough cores" in CPU's already. The claim is that software developers aren't going to be able to utilize an ever-increasing core count because they can't even keep up with 4 or 8 right now. Well, that's certainly the exact opposite of Intel's future roadmap, and Intel isn't even waiting until mass-core is physically realized before they start creating demand. They've got a vision, and they're going to see it realized. Dreamworks exec John Batter says that the key to the decision to dump AMD was that "Intel is rearchitecting [their] software tools... to take advantage of multicore and make [their] renderer highly scalable as well as making [their] character animation tools highly scalable." In short, Intel went in and invested resources to create the more advanced application to take advantage of their technology. Everyone wins (except for AMD).

nVidia does this, too, and that's why Gelsinger's recent poo-poo'ing on CUDA is misguided. He should know that better than anyone, because if Intel is going to make ray-tracing a real force, they'll eventually be doing the same thing there. Just like nVidia subsidizes games like Crysis to create demand for their high-end graphics solutions, Intel will have to do the same thing to not only prove ray-tracing's worth visually, but to give customers a reason to need it (ie, "I want to play that game and have it look like that, and I'll buy that new chip to do it").

This is how this stuff works. The solution and the demand never simultaneously appear in the tech industry. How many years has it been now since AMD produced the first 64-bit CPU? We're only just starting to get there in the consumer space to where we're really starting to have realistic 64-bit solutions and we're even starting to turn that corner where the natural demand is starting to develop as a complete ecosystem evolves. Soon, 64-bit computing will be the norm.

Consider even how many people snubbed their noses at the Personal Computer itself?! It'll be the same thing with any forward thinking technology, and any negative PR along the way is just hopeful wishes from the competition.

Continue reading >>>

Acer's Predator Gaming PC is a Big Mistake

Acer, the Taiwanese PC manufacturing behemoth, recently announced it's entry into the gaming market with the launch of the "Predator", complete with hydraulic-action front bay door, bright orange finish, and lightning bolt photoshopped background. Not to be outdone by its rival Asus' Ares desktop, Acer is following their nemesis down the same inept path that embodies everything that is wrong with the industry today.

We need MORE COW BELL! No, Alienware and every other dragon case integrator out there certainly can't have tapped out the market for gaudy computer chassis. There's always room for one more. After all, it was marketing genius 10 years ago -- it's GOT to still be good now, too. Throwing a measly few $mill$ at an over-hyped marketing campaign and a new flash website is sure to snag someone.

Support? Acer? Huh? I think it's been pretty well established that monster companies just don't adjust real well to niche markets that require a very personal, direct support experience -- especially companies based in Taiwan trying to address US markets. Further, Acer is not exactly known as the high-end, quality brand around these parts. I've talked before about support being the real service an integrator provides... that'd just about make this a slam dunk, huh.

Here's the scary part: They'll figure out how to sell more than a few of these by the sheer weight of their company. So who's going to end up with them? Customers who either 1) don't know any better, or 2) don't care. Both will end up severely disappointed, and they might just end up with a PR incident around a support fiasco that they didn't foresee, making the whole thing probably end up as a pretty negative endeavor overall. So, why bother.

Acer, prove me wrong : Show me US-based support, staffed with people who specialize in this type of hardware, with the goal and authority to provide whatever it takes for the customers who buy this thing to have a luxury-level, 100% satisfying experience.

Continue reading >>>