The Next-gen Situation discussion *spawn
- Thread starter gongo
- Start date
You guys are missing the point.
Bottom line, MS is making a lot of money right now off of xbox. And they stand to lose that (growing) profit stream if they screw up the xb720 offering. Some may choose a competing console, some may choose PC (not a huge loss, but a loss none the less), some may choose IOS/Android, and some may choose none of the above.
At the inflection point of a generational shift is not when one wants bad press extolling the underwhelming spec of the latest and greatest hardware offering(s), especially when it can be easily avoided.
What is the magic formula of spec budget vs profit? I don't know exactly what it is, but I do know what it isn't... and that's a huge risk to their currently golden goose.
Last edited by a moderator:
Some random guy on twitter has an opinion
@ID_AA_Carmack Hi John, Do you think next gen we will see more 60fps games on consoles?
@Razor81_ There will still be lots of 30hz games, which I don't think it is a good trade. If TVs didn't add lag, it would be more clear cut
Exhibit A why Carmack has lost it...
How so? I think he's right. Lots of games will target 30fps, no matter how powerful the machine. And he thinks that's too bad because of input latency, which is made worse by laggy TVs.
How do you protect that Golden Goose with hugely unprofitable hardware if you don't have another Kinect to pull out at year 5, and you can't rapidly shrink because the nodes don't exist or are too expensive?
What if you strangle that Golden Goose by making hardware you can't afford to sell and that people can't afford to buy?
Releasing a console based on phone technology isn't the answer, but neither is building another PS360. What saved the 360 this gen can't be banked on next gen, so starting from the same position would be a mistake. That's a really valid point IMO.
360 wasn't unreasonable imo and building "another 360" would be great.
ps3 was, but all because of a $200 blu ray surcharge which is a seperate matter.
360 should be nicely profitable as is, and if the unforseen one off of rrod hadn't occured, even morseo.
I dont think Kinect changed the 360's trajectory vastly. I'd have to go back and look at sales but nothing was night and day.
ps3 was, but all because of a $200 blu ray surcharge which is a seperate matter.
360 should be nicely profitable as is, and if the unforseen one off of rrod hadn't occured, even morseo.
I dont think Kinect changed the 360's trajectory vastly. I'd have to go back and look at sales but nothing was night and day.
I think Kinect has had a big impact on 360 sales, a big effect on profit margins of console sales, and a massive impact on the way the 360 is perceived by the public and used as a casual-friendly entertainment and advert box.
It would be fascinating to know how much user behaviour of the Xbox dash & services has changed since Metro and how usage varies between Kinect and none-Kinect devices.
The profits from sales of Kinect peripherals alone must be into the hundreds of millions of dollars (and after less than two years).
Really?
What is incorrect with his statement that due to the input lag & processing latency many TV's add on top of games (which is over and above user input latency in the game engine as-is) pushes the balance more toward 60Hz over 30Hz -- especially in contrast to pre-HDTV where CRTs has essentially zero latency -- isn't a real design consideration?
I know a number of gamers here on the forums who on one TV played Halo 3 fine but on another unit with more input lag and processing lag (especially pre-"gaming mode" units) had huge amounts of trouble. Moving to 60Hz tightens your game-side latency tolerances. It won't fix the display lag issues but if the user is getting 2x frames and the game has made up for some of the latency by shortening the game-side latency it tends to push the "arrow" the other way.
Btw, as an opinion, my opinion is this generation is I agree with him. Trials HD & Evolution, Forza, and the MW games have eaten a HUGE chunk of my gaming time and one big reason is they games run really smooth and are very responsive (and my display is essentially a 0 lag monitor).
I really don't know how some people do it when they have those large screen displays with poor latency and then pop in a 30Hz game with a choppy framerate. I can play games like that but it just doesn't have the same tight, responsive feeling to me.
I get a WII vibe from Kinect when it comes to long term. If it's a part of XBOX3 then there will be games and customers, but i doubt very much that it will stay fresh, and hot. Like Move and to a certain extend the Wii it did not create a new flourish of must have games, the games lack imho, actually i think they suck to the extreme, apart from the few family fun stuff games there is nothing pulling me in. And the technology just isn't good enough imho to really break any boundaries.
And the user interface with kinect is neat, and slow compared to the use of the controller. As for metro, that interface is harder to navigate than the XMB, at least from the tests i did with a 6 year old kid. He has no trouble with the XMB, and find things that he wants to play. With Metro it's less intuitive and causes more problems.
But excellent marketing, i pretty much bought it, until i saw it at a friends place, then reality took over
So you believe that MS launched xb360 with the intention of selling a $150 kinect 5 years into it to help offset the woefully expensive BOM which would have otherwise buried any profitability of the department?
Come now.
Kinect was a great boost, but MS was projecting toward profitability before the launch of the ADDITIONALLY profitable Kinect.
________________
As I said, xb360 arcade was launched at $299.
So let us assume that the GPU and CPU for xb720 are the same physical size on the latest node possible at the time. The new 28nm node wafers are reportedly 25% more expensive than previous nodes (currently).
That brings the total BOM to $390.
Now who here honestly believes MS would have some great difficulty in selling a 28nm-based xb720 (with the same CPU and GPU budget as xb360) for $399?
_________
Some here may counter with "but what about integrated kinect2?".
Sure, have a SKU with Kinect2 as well for $499.
Or, they may decide that Kinect2 is so important, they want it bundled with all consoles...
So they decide to eat the cost and bundle it for $399 ($100 loss).
Let's not forget that PS1 way back in 1995 sold for $299 and Sony lost $100 per console at that price, yet, miraculously, they were able to make a stellar profit that gen...
And that was without the xblg fees, online ad revenue, online search revenue, zune movie rentals, zune music, and potential related win8 ecosystem sales ...
______
As for the risk of future roadmap of process shrinks, if it were really that serious, intel's stock would be in the dumps. Difficult issues to overcome? Absolutely. But if intel feels confident in their roadmap, I'm sure TSMC is too.
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.
In extremely limited numbers with the vast bulk being Pro. Like 5:1 or something. And the arcade required a memory card that cost $50 and probably $5 to make (64MB!! in 2005!!). And HD cables were $40 using a proprietary connector, where as now a $5 connector from any shop will do.
Is that a launch estimate? 360 in clear profit with headroom to spare? On day one?
That looks very different to some of the other breakdowns I remember. If that's a launch estimate it smells bad. Case, fans, heatsinks, ANA, flash, southbridge, mobo, PSU, heatpipe coolers and everything else for $25? And "controller" being $"N/A" so it doesn't count? Makes you wonder why MS shit money out their ass for three years!
An analyst earlier this years pegged the 250GB 360S as making MS $115 profit at a $300 RRP. That "estimated Cost after 3 years" looks particularly funny. Could you provide a link to that analysis?
Sell it? No. Selling it profitably at $399? Yes, I could believe they'd find that difficult.
But then Kinect won't be standard.
Because the 360 was profitable from day 1, right?
Ergo, a console that launches at a $100 loss will always make a profit. Like the Saturn.
100$ loss making console always profitable; all the rest is just gravy.
You might have missed the 6 month Ivy Bridge delay that occurred over the last six months. And Global Foundries being over a year behind. And TSMC being behind on their roadmap, and also losing the next gen AMD ULP Fusions (back, lol) to Global Foundries. And Nvidia's problems with 40nm and 28nm. And Nvidia's concerns about 22nm and beyond.
And Xbox and PS3 still being on 45 nm. And WiiU being (at least partially) on 45nm. Srsly.
But it wasn't profitable from day one nor does that chart say it was. It's an analysts estimate of the BOM that doesn't include other costs such as overall testing, assembly, shipments, retail cut. It;s not a estimate of the final cost to MS per unit when sold to a customer.
It doesn't include the controllers even include the controllers, and even if you take out the HD, the BOM's price is above the $299 retail price of what the arcade bundle launched at and was the most availble SKU (no hard drive/no wifi/no wireless controller/no hd cables).
Your thesis is that Kinect "saved" the xbox division this gen and that boost can't be counted on again for nextgen.
While I agree that some magic-out-of-the-hat peripheral is likely NOT going to appear for MS or Sony 5 years post launch, I fully disagree with the notion that somehow MS risked red ink for the entirety of the xb360 product lifecycle if it were not for the Kinect.
Even if you assume $100 profit for each of the 10 million which were sold, that only nets $1B.
XBL makes roughly the same annually... and growing.
Ad revenue on top of this
Zune movies on top of this
Zune music on top of this
and (huge potential) search ad revenue on top of this.
Kinect was a nice boost, but hardly the difference between profitable or not.
I'm not arguing for a 2010 replacement... for the rest, see their quarterly statements.
Regardless. They were comfortable in taking the losses on the xb360a and figured they would make up for them elsewhere. Same for the xb360p.
7 years later and it's the fastest growing division in MS.
I didn't make the chart. It says sourced by Merrill Lynch. If you have another that you feel is more accurate, feel free to post it.
Regardless, the point is the section of that BOM which would be affected by higher 28nm costs is the GPU and CPU which they pegged at $200 total. TSMC said wafer costs are up 25% over traditional costs for 28nm (back in late 2011).
This brings the cpu+gpu budget from $200 to $250.
This additional cost can easily be shifted to the consumer upon launch. Nobody would baulk at a $349 xb720 arcade unit ($50 over cost with all else being equal).
That's the point.
How much would this silicon cost MS and what would that translate to in MSRP and sales.
Now some may say "but all else isn't equal, this costs a bit more or that". To that I say, "Correct, and in addition to those costs, MS has a significantly larger revenue and profit stream with which to offset those costs."
It was a bing image search for launch BOM. Bring any further research to the table. My point isn't on the overall BOM. The point is, "How much did the silicon budget cost then, and how much would it cost today?"
The rest is immaterial as it was acceptable then and the money being made now outside of hardware is significantly more than in 2005 and will only grow larger (if they don't screw it up).
Again, the money being made outside of hardware more than makes up for it. But if you have other BOM analysis, I'd be happy to see it.
That's MS option for how valuable they believe kinect to be. The most important demographic of a new sales generation is the hardcore fan. They influence sales of others and establish a base as well as buy a lot of games and accessories.
This demographic has so far been a difficult sell for kinect.
If MS values the kinect 2 interface and wants to see it as a standard, they need to eat the cost or hide the cost. A $100 apu doesn't do that for them.
No, but a console that sells for a profit is always the market leader which also brings billions in revenue from other channels other than hardware, like Nintendo64.
It was an example of a profitable model which worked all the way back in 1995.
It worked then with far fewer revenue streams ...
And yet, miracle of miracles, I'd bet my left testicle that some how, some way, intel will be selling 22nm chips this year. Delays happen. Especially when competition level is low.
Bottom line, process shrinks will happen, even if they get delayed. All that does is delay any potential price drop related to the node shrink or delay profits from the hardware channel.
But as I've said repeatedly, the revenue streams outside of hardware far outweigh the realistic potential of revenue streams on hardware.
Now if the hardware isn't compelling to begin with, say goodbye to not only hardware profits, but xblg profits, zune movie profits, zune music profits, game platform profits, accessory profits, ad revenue profits and search ad profits. Not to mention a weaker overall platform ecosystem with which to sell all of the above on...
But I'm sure that's worth it to save ~$100/ console ...
There are also gains to be had targeting a fixed platform like a console over a PC platform. In short you can do more with less, I.E. equivalent performance from less than equivalent hardware.
I have to go back to this because it's so great. The "Other components = $25" and "Controllers = N A" are still my favourite bits. Is this a pre-launch estimate by someone down the pub? Using these figures in preference to the more detailed and more easily found iSuppli figures just seems really strange.
Here's an iSupply launch teardown (the power supply alone is greater than the "Other components" figure provided above):
http://www.pcpro.co.uk/news/80708/isuppli-reckoning-the-xbox-bill-of-materials
iSuppli launch figures said:
This is a BOM and doesn't even include stuff like shipping and retailer margin.
And the Arcade sold for $100 less with only the omission of a $40 HDD and a few bucks for cables. Even with memory card sales the Arcade would have been losing a huge amount of money. No wonder MS didn't want you to buy it. That's a deep hole MS were digging.
Now let's canter along to 2006 and see how things were getting along:
http://www.xbitlabs.com/news/multimedia/display/20061120132150.html
iSuppli 2006 figures said:
Again, this is a BOM and doesn't include shipping and retailer margins. Depending on other variables you might be breaking even on the 360 20GB with these figures, but still losing money on the Arcade.
Of course, we know MS were still losing money because of huge repair bills, engineering costs for upgrading the system and added HDMI, engineering costs of trying to stop RRoD. A period of reduced (or ceased) manufacturing, a more expensive heatsink and a continued period of slow sales would come in early 2007.
Good job MS didn't have AMD Llano style continued yield problems.
*The PS3 was super expensive even without Bluray.
I can't believe you used such a shitty cost breakdown as the basis for your "lose an extra hundred dollars at launch!!" system, TheChefO!
Last edited by a moderator:
The iSuppli BOMs I linked to above include manufacturing costs. The Arcade sold in very limited numbers early on, and sure as hell wasn't the most available SKU. That was not an accident.
No 360 shipped with wifi until 2010.
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.
Yeah, I agree with you on the games front. I think it's about building a form of interaction with the platform for none games stuff that will create its medium term value to the Xbox (and MS). Hopefully the next generation of Kinect will allow for finger tracking and face reading, and that will be where boundaries can really get broken. Maybe. Might need an learning, intelligent interface to help bridge the gap.
Yeah again. I don't like the metro interface, and find it slow, cumbersome and annoyingly ad driven. However, some people seem to prefer it to holding a controller and scrolling through lists (I don't). It's not quite there yet, but a combination of user conditioning and interface improvement should allow it to be a partial, permanent interface method (kind of like touch and swipe vs keyboard).
But excellent marketing, i pretty much bought it, until i saw it at a friends place, then reality took over
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?
