Do you think it would be a mistake for MS and/or Sony to launch in 2012?
- Thread starter kagemaru
- Start date
So really, this thread isn't "would it be a mistake for xxx to launch 2012" but "given this set of hardware and these expectations about the software development platform and these presumptions about the business model of next-gen, would it be a mistake..."
So in that I have to say 'yes', you're absolutely right! If everything runs according to your expectations then the release date you pick is a great one. Of course this leads to cyclic arguments. Because they'll launch early, they'll want derivative hardware instead of more efficient designs (your next-gen hardware argument), and the fact they'll launch early is used to prove that they'll want derivative hardware.
What happens if those expectations aren't right? What if next-gen hardware is something completely different, like ARM and GPGPU, because it gives way better bang-per-buck? What if the programming environment takes a shift to support different paradigms (GPGPU)? What if the hardware choice wants a bleeding-edge tech like stacking for better economies, necessitating a wait until 2013?
So none of them were the "Xbox previous" straight rebuilds that you're suggesting support the Xbox 3 launch.
You seem to have formed the connection that because the 360 launch included ports, and that because ERP said that part of the port process takes/took him < 3 weeks, that it means you could cook up a launch lineup in weeks and do so for so little money that publishers wouldn't mind if the launch didn't happen and games missed marketing and were left to age till the next year.
To me this seems to be a flawed line of reasoning on many levels, and if the rumours of early dev kits being worked on in secret are true, then it would suggest that games being ported with only weeks of development time is not likely. Even WiiU ports of PS360 games are several months into development and yes, I bet publishers would be pissed if the Wu release date were to slip back to 2013.
Then, as I stated, it would be rather difficult to conform to such a timeframe given what we know at this point WRT developer involvement.
The decision isn't made in a vacuum.
There are pros and cons.
Launching in 2012 with derivative hardware (not necessarily "weak" mind you) has the advantages of potentially capturing first-mover/hardcore gamers (they like to buy the latest and greatest and spend money...it would take something special to convince them to buy another machine which does roughly the same thing within a year), mindshare for others that would like to purchase, but may not be able to afford it, being the dev base for nextgen development (MS seems to already be filling this void by establishing early NG dev kits at EA, Ubi, Crytek, Epic), games library (getting started earlier means more games which have this hardware in mind will begin, and end on said hardware) and the "friend effect".
Cons are: more expensive to launch than 2013 (assuming roughly equal hardware spec budget), potential disruption of existing xb360 base (not likely, given the target demographics at a given timeframe), potentially being outclassed by competitors hardware, thus losing the hardcore/first-mover base (which is why it would be stupid of them to come to the table with a "happy meal" for hungry gamers which could lead to a breakdown of all of the advantages listed above).
Choosing which hardware to put in is less important than choosing what the BOM budget for said hardware is IMO.
Simplified and unrealistic example of what to expect but:
In other words, lets assume the "GPU+CPU(+Ram)" die area for xb360 and ps3 was 500mm^2.
One chose to literally scale the exact same chipset, features, and functionality of their existing hardware box and produce it on 28nm with the same size 500mm^2 (literally a ps3 x8) and the other chose 250mm^2 but using whatever latest and greatest architecture.
I'd go out on a limb and say the 500mm^2 "ps24" with 4gb ram, 64SPE, 8PPE, 4mb cache, and RSX (x8) launched in 2012 would wipe the floor with 250mm^2 "Xbox Super Arm GCN edition" with stacked memory and a total of 2gb ram in 2013.
One of these will have toolsets that let developers shine with what they already have cooking and will support every game in their DLC library and even cross-generational online multiplayer. The other, won't be on the shelf for a year, will ask customers to buy all their DLC again (this time, with tesselation!), and will have mixed reviews when it comes to the visuals of the more limited library which is available.
Extreme example but it makes my point that Die Size budget is what will make or break these machines in providing an edge.
Back to reality:
By "derivative", I was speaking more from the standpoint of the CPU architectures (Sorry if that wasn't clear by now). Not that they will be exactly the same as they are now but scaled up. I expect there will be some changes, but nothing radical and nothing that would break BC.
As for GPGPU, I don't see a reason why either machine couldn't have GCN while still having BC. GCN will be available commercially in q1/2012. Way back in 2004, MS had ATI put together a chip which utilized tech ATI wouldn't have at retail for 2 more years. GCN should also not have a significant problem in replicating or emulating Xenos or RSX.
So I'm not seeing a reason for GCN to not be in a 2012 console. It will be readily available, flexible, affordable, and compatible.
As for Ram, it's not as though there aren't options ~today that would provide the bandwidth necessary for a nextgen console. GDDR5, XDR2, EDRAM+DDR3 (ala xb360).
_____________________________________
Making the move in 2012 would not preclude some special sauce technology (that I'm aware of) that will be commercially viable in 2013 that isn't available for 2012.
If there is and it would provide a sizable jump in either equal performance for half cost (lifetime not just teething problems at launch) or double performance for same cost, I'm all ears/eyes.
It would have to be significant (perhaps not as extreme as illustrated above) to offset the loss of a year on the market (and all that entails), and potential loss of BC.
Last edited by a moderator:
No, it's exactly what it should be titled.
Would it be a mistake?
Well here are the pros and cons...()
What could be done in 2012?
What could be done in 2013?
What are the implications of choosing x over y and when would these technologies be available? etc.
My point of view is that there is little on the near-term horizon which would be a significant game changer to warrant waiting.
I've outlined why I feel that way and why I feel 2012 is a viable window if Sony/MS chose to dive in.
The process technology is there to make a significant jump over xb360/ps3 for a reasonable cost (barring price gouging by TSMC).
Now if Sony/MS choose to wait, it would be for other reasons (choosing short term profits over longterm, knowing competitors timeframe isn't a threat, Nintendo Wuu will flop, and Google/Apple would never eye the lucrative livingroom market ... nor think of swooping in and stealing their lunch
).I've suggested 2012 wouldn't be a mistake for the reasons I've outlined.
But that's me.
Xbox toolsets allow for a pretty straight forward port between PC and xbox360.
Yes?
PC versions of xb360 games have higher res rendering, AA, textures, filtering, and/or adjusted LOD bias, yes?
PC games are now starting to utilize DX11 features, yes?
Now if we are to assume (yes we all know what they say about assumptions) that the architectures for NG will be derivative, then the toolkits can be applied (with alterations) relatively quickly to enable cross platform development between PC and xb720 (or ps4).
As for why EA and the like have had dev kits for a while... Probably to give them time to get familiar with exactly what the new box can do, get to work on preliminary port/cross-development work of expected titles for the launch window of the titles on both xb360 and xb720, and to start work on NEW titles/IP which utilize the spec from the ground up.
Year 2 or 3 is when software finally starts showing up that taps the capabilities of the systems, and that work needs to start sometime.
The more familiar the hardware is, the sooner devs can reach full potential.
As for the original xb360 launch lineup, I imagine much of the time it would have taken would be due to the very unfamiliar hardware. Tack onto this the unfinished toolkits MS had at the time, and the fact that devs didn't even get final xb360 hardware until a few months prior to launch. Such a chaotic mess will not take place if derivative hardware is in place and the assumed launch window is q4/2012.
What actual effort went into the xb launch lineup I don't know (aside from the fact that the models, textures, and games were roughly identical to those on xb1/pc), but I recognize it was not an ideal environment.
This time around, it could be.
Last edited by a moderator:
I know what the pros and cons are - you've repeated them in every thread you post in!
I'm just wondering when you'll agree to disagree and stop posting, instead of repeating yourself over and over expecting something to change in the discussion.I'll also say the thread is titled wrong, as it's not an open question. Or rather, you started the wrong thread. Really, this thread (and all your posts) should be consolidated into a "TheChefO's Strategy for Next Gen" thread. There you'd outline your whole argument instead of spreading it piecemeal over several threads, wherein it's obvious to the participants that there are several dependencies within your strategy that need to be executed in parallel for it to work. You say, "I think MS should launch in 2012 using derivative CPUs, high-level GPU calls, and existing XNA tools, clearly explaining to devs that their current games in development will be forward compatible as long as they adhere to some XNA fundamentals. MS'll gain first-launch advantage giving blah bah bonuses, while Sony launching later won't benefit greatly from the possibilites of more mature tech because of diminishing returns. Forward/backward compatibility will give MS the chance to continue the 360 line as a set-top box and cheap Kinect device."
Everyone would then respond, "Yeah, that sounds like a valid business strategy. It is/is not what I'd do. I think MS will/won't do this." We then avoid all the fractured debates of you talking about hardware in one thread that depends on a launch window, and a launch window in another thread that depends on hardware choices, and the cost of game development in another thread that depends on assumptions of developer environments and code portability.
I know what the pros and cons are - you've repeated them in every thread you post in!
I'll also say the thread is titled wrong, as it's not an open question. Or rather, you started the wrong thread. Really, this thread (and all your posts) should be consolidated into a "TheChefO's Strategy for Next Gen" thread. There you'd outline your whole argument instead of spreading it piecemeal over several threads, wherein it's obvious to the participants that there are several dependencies within your strategy that need to be executed in parallel for it to work. You say, "I think MS should launch in 2012 using derivative CPUs, high-level GPU calls, and existing XNA tools, clearly explaining to devs that their current games in development will be forward compatible as long as they adhere to some XNA fundamentals. MS'll gain first-launch advantage giving blah bah bonuses, while Sony launching later won't benefit greatly from the possibilites of more mature tech because of diminishing returns. Forward/backward compatibility will give MS the chance to continue the 360 line as a set-top box and cheap Kinect device."
Everyone would then respond, "Yeah, that sounds like a valid business strategy. It is/is not what I'd do. I think MS will/won't do this." We then avoid all the fractured debates of you talking about hardware in one thread that depends on a launch window, and a launch window in another thread that depends on hardware choices, and the cost of game development in another thread that depends on assumptions of developer environments and code portability.
touché!
For some reason, people seem to forget the shitty economy basically worldwide. I don't see the appeal of launching a pricy $400+ console in 2012 when whoever launches in 2013 will face a far better economic climate.
I feel like this is the strongest reason why due to the economic collapse in 2008 that dragged on for years, Microsoft didn't even attempt to create a console to launch in 2011 (6 years after the Xbox 360s launch).
I don't really believe the hype that Kinect has "extended" the life of the Xbox 360 because by the end of 2010, Microsoft should have had a good idea whether 2012 or 2013 was their primary target. Kinect selling 8 million units shouldn't have gotten MS to think "that sold so well, so lets delay our next console by a year!!" otherwise, I seriously question their judgement.
I feel like this is the strongest reason why due to the economic collapse in 2008 that dragged on for years, Microsoft didn't even attempt to create a console to launch in 2011 (6 years after the Xbox 360s launch).
I don't really believe the hype that Kinect has "extended" the life of the Xbox 360 because by the end of 2010, Microsoft should have had a good idea whether 2012 or 2013 was their primary target. Kinect selling 8 million units shouldn't have gotten MS to think "that sold so well, so lets delay our next console by a year!!" otherwise, I seriously question their judgement.