[hugh]

Microsoft is debating whether the next Xbox has to be profitable on a per-unit basis from the day it's released, or whether the company can afford huge subsidies as it did with the first two consoles.

According to insiders quoted by game industry publication Kotaku, Microsoft's board of directors has taken up the argument as Microsoft begins to plan the next generation Xbox, which will probably come out in 2014 or 2015.

Selling game consoles for a profit from day one would be a huge strategic shift.

 

[Proteosome]

Considering the product lifecycle of these consoles it would seem to make more sense to initially sell at a loss to maximize market share. Look to the leap in market share the xbox attained in this generation compared to the ps3 (also sold at a loss but at a much higher price point).

 

[99semaj]

Problem is, one can't have complete control over the life expectancy of the product. In this case, they are squirming a little bit because Wii is about to raise the bar to some extent, and PS3 has a lot more headroom left than XBOX and it's only midway through the planned cycle. That changes the economics of the business case they pitched five years ago.

I think the BoD is likely trying to instill greater discipline around the business case because the program has done little more than buy marketshare...while it's recently become profitable on a go-forward basis, the entire program may never be in the black, and they have ten years experience at this now...it's time to act like a mature business.

I think the end result will be a compromise between the BoD and the program leadership. It's reasonable to launch in a loss position so long as you have strong operational management to manage deflation, value-engineering and cost-out efforts.

 

[audacity]

and PS3 has a lot more headroom left than XBOX

Not according to game developers that I've met. Specifically they talked about how they need to use lower resolution textures on the PS3 when compared to the PC and XBOX because of the segmented memory architecture found on the PS3.

They also mentioned that due to Microsoft having indisputably better developer tools, games are usually developed using the Microsoft tools and then later ported to other platforms. It's not surprising that Microsoft has better development tools than Sony since software has never really been Sony's forte. I only bring it up because during these console X vs console Y discussions, without fail, someone comes along and says "but what really matters are the games". The best thing that you can do to arrive with better games is to make developers lives easier by allowing them to work on the the problem of building their game rather than having to fight with crappy developer tools and ridiculous asymmetric processor architectures.

Oh, and if you're curious, the developer I met that was telling me how their games use lower resolution textures on the PS3 worked for Ubisoft Montreal, and he was specifically working on Assassins Creed: Brotherhood at the time.

 

[audacity]

From the article:

The risk: if Microsoft goes the Nintendo route, Sony might be willing to accept huge per-unit losses to come out with a more technically advanced console and reclaim the lead.

I'm not sure the correlation between most advanced console and "the lead" is clear. The original XBOX was much more powerful than the PS2, which sold far more. The XBOX, XBOX 360, and PS3 are all more powerful than the Wii, which sold far more.

The N64 was more powerful (from a rendering perspective) than the PS1, etcetera.

 

[eimaj]

Not according to game developers that I've met. Specifically they talked about how they need to use lower resolution textures on the PS3 when compared to the PC and XBOX because of the segmented memory architecture found on the PS3.

From a third party perspective. First party developers such as Naughty Dog and Guerrilla do not seem to have any issues with this.

 

[audacity]

From a third party perspective. First party developers such as Naughty Dog and Guerrilla do not seem to have any issues with this.

I'm not saying the PS3 is horrible, I'm just saying that it often suffers from lower resolution textures than the 360 due to hardware design choices. This is a big win for Microsoft because they were able to:

1. In the world of moore's law, produce a console that has similar effective hardware capabilities to the PS3, and release it a year before the PS3 was released.

2. Relegate Sony - the dominant player for the previous two console generations - to last place.

3. Start turning a profit.

Now Microsoft is over the hump: the business group containing Xbox has turned an operating profit for each of the last 11 quarters, including a $1.05 billion profit last quarter thanks to sales of Kinect.

Without question, this current generation the big winner is Nintendo, and the big loser is Sony when considering their fortunes from the PS2/XBOX/GC era.

Anyway, I just wanted to correct 99semaj's comment on hardware headroom because I get the sense that his seething hatred for Microsoft exceeds his technical acumen.

 

[99semaj]

In the interest of keeping this on topic and not feeding the trolls, please allow me to voluntarily withdraw the comment about headroom. The topic per the OP is the business side of the equation, not the highly subjective fanboy discussions that inevitably hijack an otherwise interesting thread

The BofD for MSFT have posed the question because they are obviously trying to avoid further losses for the business segment. They have already spents billions on buying share, now they want to realize a return on that investment in the next phase. The program leaders would be making a career limiting move to ignore this shot over their bow.

 

[jvincent]

Assuming the Xbox720 (or whatever it's called) has a similar architecture to the 360 in terms of how it's programmed I can't see why it would need to be a loss leader.

AFAIK, the 360 is uses essentially the same DirectX graphics engines as PC games.

If that's the case the next generation simply needs to push the envelope on the silicon technology and they get the following:

1. Backward compatibility for all current games
2. Processing power that is many times greater than the current box for new games

The big cost of the original was in building the eco-system but that is done now.

Assuming a 2014 platform launch they will be looking at 22nm, or maybe 18nm for the chips. They won't be cheap, but they should not lose money.

 

[audacity]

Well, just because a game uses DirectX doesn't make it easy for backward comparability. The 360 did a reasonable job emulating the original XBOX in software, but that was using Rosetta-style binary translation/emulation system. If they stick with a PowerPC processor for their next console then it shouldn't be that difficult to maintain backwards compatibility, but if the go to a new architecture again it will probably be hit-and-miss.

As an interesting point of history/trivia, the x86 to PPC translation engine they were using for the XBOX 360 to play XBOX games originally came from the "Virtual PC" software that Microsoft acquired from Connectix that was an important product for Mac users to run PC apps back in the day. I learned this from seeing a virtualization talk given by a Microsoft employee who used to be a Connectix employee and wrote the code in question.

I think they may hop chip architectures again because I have a hard time believing that the PPC instruction set is where you want to be right now, especially since there are many interesting products being developed by Intel and AMD that would be excellent console chips. Specifically, I'm referring to the Fusion products from AMD, and Intel's planned enhancements to their on-die GPUs. Sure, Intel's GPUs suck today, but I suspect they will be a lot better in 3 years.

These Fusion chips are especially attractive if the next generation consoles are going to target a lower price point instead of trying to push the performance envelope.

 

[jvincent]

True enough, I guess my point was that if they choose to adopt a new S/W architecture then they are taking on a world of hurt. Free advice to MSFT, don't do that!

But unless IBM drops the Power architecture completely I don't see MSFT changing to a different platform since it would be crazy to force the developers onto a new S/W platform.

I don't think the on die-GPU products aren't really an option unless as you say cost is the new benchmark. Even if it were, it's relatively simple to take the current chipset and smush (technical term) it onto a single die at a smaller geometry to produce a lower cost 360.

My guess is that for the new console they are going to up the rendering ante significantly and that takes silicon area.

 

[asif9t9]

Would there be a way to go next gen without a full console, sort of like Onlive? I figure that's how MS could save money to start, by not investing in so much hardware.

 

[james99]

Getting to market first for a decent price is the key, even if you're losing money.

I have all 3 generations of the PS and the 360. I don't own the original Xbox.

I only picked up the 360 because it came out a year before the PS3. Otherwise I would have stayed in the Sony only camp.