The Next-gen Situation discussion *spawn

[function]

Huh? Arcade models were certainly never supply constrained, the pro models most certainly were at launch and yes the launch 360 didn't include wifi. Arcade models also didn't ship with gold membership, a plug and play charger/battery pack or a wireless controller which were included in the pro.

The Core and the Arcade were different. [Edit - I'm guilty of using "Arcade" to describe Arcade &/or Core myself, so I'm not trying to patronising]

I remember retailer comments around launch that they were getting a lot more 20GB systems than Cores. The shops I went in to in the UK confirmed this - Game, Gamestation and the Indies.

20GB models were far more in demand because they weren't shitty value like the Core. Once you'd bought a memory card and in some cases a HD cable you were almost up to 20GB prices. MS didn't want you to buy the Core because they lost more money on it. But if you refused to be coerced into a 20GB they didn't want to lose you as a customer, and crucially the Core was the price they got to use on all the adverts and that was what got you into the store (to be coerced into buying the 20GB).

The wireless controller cost almost nothing extra beyond the wired version, but was again there to nudge you away from the Core and towards the 20GB. Once the Core became cheaper to make - after the Falcon was introduced - it magically turned into "Arcade" and got the wireless pad (surprise, surprise) and also came bundled with a 256MB memory card (attractor) so you no longer got stiffed on buying an expensive and shitty 64MB memory card (repulsor). Effectively it was a facelift and pricecut in one, and the Arcade actually became an attractive buy.

You can see the exact point when the role and placement of the of "Core" changed and that was when the "Arcade" appeared. The ugly duckling turned into a beautiful (and cheap) swan.

Gold membership was a free month and didn't add to the BOM. No 360 shipped with a plug and play wireless charger afaik.

 

[Shifty Geezer]

I did buy it! But I hardly use it. I enjoy the idea of Kinect more than actually using it.

I don't know what that means?

It means their marketing department was very good.

 

[Brad Grenz]

The Core and the Arcade were different.

I remember retailer comments around launch that they were getting a lot more 20GB systems than Cores. The shops I went in to in the UK confirmed this - Game, Gamestation and the Indies.

20GB models were far more in demand because they weren't shitty value like the Core. Once you'd bought a memory card and in some cases a HD cable you were almost up to 20GB prices. MS didn't want you to buy the Core because they lost more money on it. But if you refused to be coerced into a 20GB they didn't want to lose you as a customer, and crucially the Core was the price they got to use on all the adverts and that was what got you into the store (to be coerced into buying the 20GB).

The wireless controller cost almost nothing extra beyond the wired version, but was again there to nudge you away from the Core and towards the 20GB. Once the Core became cheaper to make - after the Falcon was introduced - it magically turned into "Arcade" and got the wireless pad (surprise, surprise) and also came bundled with a 256MB memory card (attractor) so you no longer got stiffed on buying an expensive and shitty 64MB memory card (repulsor). Effectively it was a facelift and pricecut in one, and the Arcade actually became an attractive buy.

You can see the exact point when the role and placement of the of "Core" changed and that was when the "Arcade" appeared. The ugly duckling turned into a beautiful (and cheap) swan.

Gold membership was a free month and didn't add to the BOM. No 360 shipped with a plug and play wireless charger afaik.

Oh, yeah. The original Core existed for PR reasons and little else.

 

[function]

It means their marketing department was very good.

I bet I'm going to buy the next gen Kinect without playing it first too...

... I hope it looks even more like Wall-E!

 

[Rangers]

wat?
Could you share these projections? Because from the financials MS bled billions into the first 3 years (and that's not including 3 years of R&D). With a replacement in 2010 MS would have been even further in the red than after Xbox 1.

360 division has been profitable literally for years and years. I think 5 years, but at least 4. That's with RROD throwing a huge wrench in things.

Chef is right the trends were well established before Kinect, both for profitability and sales. I dont think Kinect is even selling that great any more, certainly the shipments have been declining.

 

[steviep]

There is no "360 division"

There is an entertainment division that includes a whole host of other things.

 

[function]

360 division has been profitable literally for years and years. I think 5 years, but at least 4. That's with RROD throwing a huge wrench in things.

Just because you break into profit in a financial year doesn't mean your previous multi-billion dollar losses are wiped out. With a five year replacement cycle the 360 would probably have been deep in the red. Sales in late 2009 / early 2010 weren't even that great.

It's the long product cycle, the ability to rapidly diminish costs, and Kinect that appear to have made the 360 into a cash cow. Taking the fact that these things have happened this generation and using them as the basis for making a similarly large or even larger (in the case of TheChefO) upfront loss next generation is what I see as being deeply unwise.

Xbox 3 may not have 8 years of being the primary core-gamer system (at a time of great popularity for standalone consoles) to recoup losses and (hopefully) make profits. It may not be able to drop 40% of it's manufacturing cost in the first 12 months, and 75% (or more) over 6 or 7 years. It may not get a multi-billion dollar boost in revenues from a Kinect like device after 5 years.

There may be a business case for losing lots of money upfront again, but doing it just because "the Xbox did it and things (probably) worked out okay" seems a bit nuts.

 

[-tkf-]

I did buy it! But I hardly use it. I enjoy the idea of Kinect more than actually using it.

I don't know what that means?

What i meant was, i bought the marketing, i bought the XBOX users glowing reports, i bought the freaking hype. Until i saw some kids using it, "step back" , "more light" etc etc.. in real life

Good thing is kids don't care about that stuff... And good thing Microsoft made a Starwars edition.. which made me a 360 owner, and kinect owner.. Difference is, that i think kinect is fun for kids that actually don't really care about the game as such, they just have "fun". Starwars for example, you just try and try and try until you progress. So the limitations of Kinect is not a great problem for a kid that just think it looks cool and have no problem fighting the user interface.

If i werent for the kids, kinect would not be plugged in, it also slows down the boot process

 

[function]

What i meant was, i bought the marketing, i bought the XBOX users glowing reports, i bought the freaking hype. Until i saw some kids using it, "step back" , "more light" etc etc.. in real life

Good thing is kids don't care about that stuff... And good thing Microsoft made a Starwars edition.. which made me a 360 owner, and kinect owner.. Difference is, that i think kinect is fun for kids that actually don't really care about the game as such, they just have "fun". Starwars for example, you just try and try and try until you progress. So the limitations of Kinect is not a great problem for a kid that just think it looks cool and have no problem fighting the user interface.

If i werent for the kids, kinect would not be plugged in, it also slows down the boot process

I didn't notice the slowed boot process, although I did notice that Kinect took a while to be ready for use once the dash was up and running (so I'd normally have finished my navigation using the pad by the time it was ready).

What I did find was that while single player was a bit rubbish, in group scenarios Kinect was even more fun than Wii Sports. My friend's kids love it and I've actually found myself enjoying playing against inexperienced gamers in a way I've not before. Flapping around like an idiot is fun in a way that I can't quite equate to controller based games - just so long as you've got a competitive partner is crime.

The user interface definitely isn't there yet though (for games or the dash) so the full potential is a long way from being delivered on. Movement is in many cases so personal that unless you just want to be an awkward virtual cursor the interface needs to learn about you and your particular way of moving. I have no idea how you build that into an API that programmers could write games for though.

 

[TheChefO]

... and using them as the basis for making a similarly large or even larger (in the case of TheChefO) upfront loss next generation is what I see as being deeply unwise...

I'm not sure if you're being intentionally obtuse or if I'm speaking another language...

Simple math here:

xb360 arcade/core unit cost = x
xb720 arcade/core unit cost = x + (25% additional wafer cost of GPU+CPU)

xb360 arcade/core retail = $299 (losses are what they are upfront)
xb720 arcade/core retail = $399

What happens in the above scenario is LESS costly than what MS had in 2005 launching the xb360. The only way it would be more costly is if the 25% additional wafer cost added up to more than $100 which would put the cost for those two chips at over $400.

Now unless you're making the case that the cpu+gpu budget for xb360 was over $400, then the plan that I have laid out will SAVE MS money vs xb360.

___________

Now let's take a look at non-hardware revenue:

2million users in 2005.
35 million now 2011.
Projected 40+million in 2012.

Based on this article from Bloomberg in 2010 (sorry, I don't have an isupply one...) xblg members are roughly half the total of members. And sales of movies, tv shows, etc. are roughly equal to the revenue of xblg membership fees.

With the above information applied to 2005's 2 million xbl users (no xbl zune yet, but all of them paying members), we get $100million/yr.

Fast forward to last years 35million users, with the same Bloomberg formula (roughly $50/member) we get $1.75B... annually.

This is without the additional potential revenue from advertising.

How big is this potential ad revenue?

Let's investigate:

___________________

The NFL makes $3B per year in ad revenue on the back of 19.8million viewers.

These 20 million viewers watched 37 events lasting an average of 3 hours each for a total of 111 hours per year.

Xbox Live's 35million members spend 2.1billion hours per month... or 60hours per person, per month ... or 720 hours per year.

How serendipitous.

So, 20 million people watching 111 hours per year generates $3Billion for the NFL.
How much do you think the 35+ million people watching 720 hours per year could generate for MS?
____________________

But no biggie, let's risk this expanding 35+million viewership by putting in a chipset which won't generate enough interest to draw sales even equal to the existing userbase, but hey, at least they aren't losing any money per box right?


Long-term thought here. The additional 25% increase on the cpu and gpu budget is MORE than made up for by the potential moving forward.

The risk of rejection from swapping in an underpowered (but profitable from day one) apu is not even close to being worth the potential reward.

 

[joker454]

What i meant was, i bought the marketing, i bought the XBOX users glowing reports, i bought the freaking hype. Until i saw some kids using it, "step back" , "more light" etc etc.. in real life

Good thing is kids don't care about that stuff... And good thing Microsoft made a Starwars edition.. which made me a 360 owner, and kinect owner.. Difference is, that i think kinect is fun for kids that actually don't really care about the game as such, they just have "fun". Starwars for example, you just try and try and try until you progress. So the limitations of Kinect is not a great problem for a kid that just think it looks cool and have no problem fighting the user interface.

You were never the intended target for Kinect, casuals/kids were. So in that respect the device overwhelmingly succeeded in meeting it's target. Guys like me who don't view consoles as core gaming anymore also find Kinect useful since while the consoles became too dated for core gaming, they are still good to me for Kinect/Arcade/Indie gaming. In that respect Kinect helped extend the life of my 360, meaning it also succeeded in courting the hybrid casual/core gamer like me (pc for core, 360 for casual). Overall Kinect was pretty spot on at meeting it's audience I'd say. Of couse the hardcore gamer will hate it, it was never intended for them.

If i werent for the kids, kinect would not be plugged in, it also slows down the boot process

It doesn't affect the boot process, you an do anything you want while Kinect continues to boot. If your machine is stalling waiting on Kinect then you have defective hardware alas.

 

[function]

I'm not sure if you're being intentionally obtuse or if I'm speaking another language...

Simple math here:

xb360 arcade/core unit cost = x
xb720 arcade/core unit cost = x + (25% additional wafer cost of GPU+CPU)

xb360 arcade/core retail = $299 (losses are what they are upfront)
xb720 arcade/core retail = $399

What happens in the above scenario is LESS costly than what MS had in 2005 launching the xb360. The only way it would be more costly is if the 25% additional wafer cost added up to more than $100 which would put the cost for those two chips at over $400.

Now unless you're making the case that the cpu+gpu budget for xb360 was over $400, then the plan that I have laid out will SAVE MS money vs xb360.

Right, so you're advocating a bigger spend on hardware for the next Xbox and a more expensive launch price to offset that. That's fair enough. The earlier bit where you said something to the effect of "or just bundle Kinect 2 and lose another $100" was what I was referrring to with "extra losses".

My preferred approach would probably be try and launch on an established process with good yields (so 32nm / 28 nm next year and not try to be the first onto 22nm) and after a decent period of design and testing. You could have reasonably sized chips (you get your 200+ mm^2 CPU and GPU) and you know everything about how they behave and how to prevent RRoD and YLoD and whatever. Basically you start at the 2006 iSupli figures instead of the 2005 figures.

Not doing another 360 doesn't mean you have to do a Nintendo and start at the 360's 2012 BOM.

Now let's take a look at non-hardware revenue:

2million users in 2005.
35 million now 2011.
Projected 40+million in 2012.

Based on this article from Bloomberg in 2010 (sorry, I don't have an isupply one...) xblg members are roughly half the total of members. And sales of movies, tv shows, etc. are roughly equal to the revenue of xblg membership fees.

With the above information applied to 2005's 2 million xbl users (no xbl zune yet, but all of them paying members), we get $100million/yr.

Fast forward to last years 35million users, with the same Bloomberg formula (roughly $50/member) we get $1.75B... annually.

We know from MS that they get less that half of that from Gold subs. I don't recall ever paying more than 75% of the rrp for a year of Live, and the last couple of years its been much less than that.

And come on, the iSuppli figures are ten thousand million times better than the ones you used. I bet even you're laughing at those figures a bit now.

"$25 misc." is going down on every price estimate I ever have to do from now on, even though I live in the UK.

This is without the additional potential revenue from advertising.

Revenue for the entire Live service was less than 1.75B iirc, though I expect it's continuing to grow.

The NFL makes $3B per year in ad revenue on the back of 19.8million viewers.

These 20 million viewers watched 37 events lasting an average of 3 hours each for a total of 111 hours per year.

Xbox Live's 35million members spend 2.1billion hours per month... or 60hours per person, per month ... or 720 hours per year.

How serendipitous.

So, 20 million people watching 111 hours per year generates $3Billion for the NFL.
How much do you think the 35+ million people watching 720 hours per year could generate for MS?

Games and movies aren't so easy to break up with regular adverts (without driving people away from your service entirely), and I doubt all tv attracts as much advertising revenue per customer as the Superbowl does.

MS have confirmed that they're looking for additional ways to stuff even more adverts in front of Xbox Live customers though.

But no biggie, let's risk this expanding 35+million viewership by putting in a chipset which won't generate enough interest to draw sales even equal to the existing userbase, but hey, at least they aren't losing any money per box right?

Most of the Xbox Live viewers you're talking about "jumped in" when the Xbox was not cutting edge and a $99 PC graphics card could whoop its ass. Practically none of the people who play on Xbox Live - and certainly none of the people using Netflicks (or watching the Superbowl) - give a damn about how new the process node was when the first Xbox rolled off the line.

Long-term thought here. The additional 25% increase on the cpu and gpu budget is MORE than made up for by the potential moving forward.

It might. Or it might just be millstone around the platform's neck when people move to their smart tvs and tablets for everything. Or when Sony release PS4.5 in 4 years and it's profitable from day one and kicks you systems ass even though your system hasn't paid itself off.

The risk of rejection from swapping in an underpowered (but profitable from day one) apu is not even close to being worth the potential reward.

The risk of being unable to lower the retail price of the machine to level that's competitive in the market place is also a big one. However much you spend on making a cutting edge machine it'll be outdated in a few years if someone new comes along.

If you need to compete on cost instead of (or as well as) features like the 3DS did, but you can't because FIVE HUNDRED AND NINETY NINE US DOLLARS then you're risking rejection every bit as much as if you'd released an underpowered machine.

 

[Shifty Geezer]

Now let's take a look at non-hardware revenue:

2million users in 2005.
35 million now 2011.
Projected 40+million in 2012.
...
How big is this potential ad revenue?

You make some reasonable arguments, but the bigger picture still paints lots of more difficult choices as Function goes into some detail. But in simple terms, if you are losing $100 per console and sell 15 million a year, that's $1.5 billion in lost profits. If you could sell the same number of consoles with a $100 lower BOM, that's 1.5 billion a year more profit you're making. Yes, adverts and alternative revenue streams would help fund a more expensive box, but these companies are all about maximising profits and not serving consumers. It would have to be proven that losing more per unit would result in more profit generated overall. That's a very difficult point to prove. The appeal of better hardware might not have enough traction in the market place depending on what else is going on.

eg. MS invest heavily in XB3 producing the most powerful console and sell it for $400 at a $100 loss to them. Sony instead release a cheap box at cost for $200. The XB3 outclasses the PS4 in peak quality, but to Joe Gamer they are both significant upgrades compared to what they are used to on their current gen machines, and they can afford the PS4. So the PS4 outsells the XB3 4:1, gains momentum, and then in 3 years Sony introduce the PS5 also at $200 that's hardware compatible offering an upgraded experience for PS4 owners on their existing games while decreasing the value of the now $200 XB3.

That's one scenario. There are many scenarios that could play out depending on who does what and when, and how well they execute. Given all these variables, how do you prove that the most expensive, lossy, powerful box is the most profitable option? For every historical example of a powerful console being successful, there is at least another example to the contrary (like Xbox, 3DO, PCE), so there are no cut-and-dried answers.

 

[Gubbi]

eg. MS invest heavily in XB3 producing the most powerful console and sell it for $400 at a $100 loss to them. Sony instead release a cheap box at cost for $200. The XB3 outclasses the PS4 in peak quality, but to Joe Gamer they are both significant upgrades compared to what they are used to on their current gen machines, and they can afford the PS4. So the PS4 outsells the XB3 4:1, gains momentum, and then in 3 years Sony introduce the PS5 also at $200 that's hardware compatible offering an upgraded experience for PS4 owners on their existing games while decreasing the value of the now $200 XB3.

A cheap box won't outsell the expensive one 4:1 unless you have a Wii-like gimmick. A $200 box would be a miniscule upgrade over current gen hardware. It would receive a tonnes of bad press because of lack luster performance. Even if you gain momentum you would kill it stone dead with your three year generation length. Would developers produce games for the large install base with crappy performance, or the small install base with decent performance ?

You end up with a box with a very short lifespan, that is hardly an upgrade, with zero userbase and very small game library.

It was noted in another thread that only 11% of console users would pay a premium for a high performance console. Those 11% are the early adopters you need to build your initial userbase and drive down costs.

I don't expect MS or Sony to go costs-be-damned like Sony did with the PS3, but I do expect a loss leader with a BOM $100 higher than selling price for the next gen.

Cheers

 

[Shifty Geezer]

A cheap box won't outsell the expensive one 4:1 unless you have a Wii-like gimmick.

There may be other factors. Sure, people can try and argue for and against every possible contingent to try and guess what the market will do, but it'll be guesses based on very little real information. No strategy can be proven the best one.

It was noted in another thread that only 11% of console users would pay a premium for a high performance console. Those 11% are the early adopters you need to build your initial userbase and drive down costs.

Except if the console launched at $200, you wouldn't be limited to just those willing to spend more. You may find that the early adopters of a new, $200 console would unmber the tens of million, some 50% of those polled.

I don't expect MS or Sony to go costs-be-damned like Sony did with the PS3, but I do expect a loss leader with a BOM $100 higher than selling price for the next gen.

I have no idea what to expect. My only concern with the $100 loss-leader is how quickly they could get it to break-even. With die shrinks yielding less returns, launching at a loss might see the hardware stuck at a loss for far too long and lose way too much money.

 

[Gubbi]

There may be other factors. Sure, people can try and argue for and against every possible contingent to try and guess what the market will do, but it'll be guesses based on very little real information. No strategy can be proven the best one.

The loss leading model has been succesful numerous times. Wii is the only counter example, which IMO will be a very hard act to follow.

Except if the console launched at $200, you wouldn't be limited to just those willing to spend more. You may find that the early adopters of a new, $200 console would unmber the tens of million, some 50% of those polled.

But a box with a cost of $200 would be a very modest upgrade. Your new mediocre platform is up against established platforms with a vast amount of users (which matters for multi player) and a vast game library.

Consider this: A loss leading box with a BOM of $300 sold for $200 vs a box with a BOM of $200 sold for $200, which would sell the most ?

I have no idea what to expect. My only concern with the $100 loss-leader is how quickly they could get it to break-even. With die shrinks yielding less returns, launching at a loss might see the hardware stuck at a loss for far too long and lose way too much money.

You need to consider all the revenue streams when you talk about break-even. A loss on the hardware can be sustained as long as they have revenue form every software, media or service sold on the platform.

As for silicon cost reductions, even if the new, cutting edge, nodes won't see immidiate cost reductions, the older will. The bulk of silicon production costs are capital costs, as the equipment is amortized production costs are reduced.

Cheers

 

[Shifty Geezer]

Consider this: A loss leading box with a BOM of $300 sold for $200 vs a box with a BOM of $200 sold for $200, which would sell the most ?

All things being equal, the $300 box. But it's never that straightforward. Maybe that hardware is expensive but not terribly efficient and the $200 box holds its own (comparable to XB360 versus PS3)? Maybe the $200 box launches a year earlier and sells so fast it seals next-gen? Maybe the $200 box comes with such an amazing user experience that it wins all the street cred and interest? Maybe the $200 box is hardware compatible with this gen and renders the same existing library in better quality? Perhaps the $200 box is sold as an all-in-one entertainment service, adding a subscription to unlimited movie and music streaming from whatever service on your mobile device and PC as well as the console?

$100 spent on better hardware could be spent on better something else. Or kept as profits if you can still sell about the same despite spending less. It is definitely not a guaranteed better investment to spend on hardware.

 

[czekon]

But in simple terms, if you are losing $100 per console and sell 15 million a year, that's $1.5 billion in lost profits.

but ms charges for xbox live (gold) so this initial 100$ loss is coming back to you anyway. And i dont think new xbl will be free. In that way they can afford even bigger cut in price (??)

 

[Shifty Geezer]

MS is getting those Live! subcriptions anyway, so you're not getting the money back. You lose $100 per console if you sell the console for $100 less than it costs to make and sell. You lose profits. Wahtever money MS would make from Live! subscription would be profits instead of deferring costs if the hardware is cheaper.

The only argument in favour of losing profits to a more expensive BOM is if by spending more, you end up making more net profits then if you didn't, by selling more consoles than you would with cheaper hardware or by that higher BOM resulting in more content sales. eg. Including an HDD gets you content purchases in excess of the cost of including the HDD, where not including the HDd would lose you that sales.

I don't think it's possible to prove that $100 more on BOM always results in more than $100 extra revenue per customer or >33% more sales or whatever it'd need to be to offset the losses.

 

[Gubbi]

MS is getting those Live! subcriptions anyway, so you're not getting the money back. You lose $100 per console if you sell the console for $100 less than it costs to make and sell. You lose profits. Wahtever money MS would make from Live! subscription would be profits instead of deferring costs if the hardware is cheaper.

If Live subscribers defect to a competing, more performant, platform, they lose the revenue. The more future revenue, the more viable a loss leading strategy is.

I don't think it's possible to prove that $100 more on BOM always results in more than $100 extra revenue per customer

It would depend on the absolute price. Cutting $100 off a $1000 box wouldn't matter as much as cutting $100 off a $400 box. There is a maximum console buyers will pay, and there is a minimum where the hardware itself won't be considered a valid upgrade.

Cheers