Two big lessons that Sony has learned from launching PS3.
1) Time to market: Game console has become a congested market. The delay launching PS3 was a big deal for Sony to timely making the development cost of PS3.
2) Game availability and time to market: Cell processor was superior in many way. The graphic quality and performance were incomparable in the early time of PS3. With the same game available in both PS3 and XBox 360, the one on PS3 was so much better. However, gamers are willing to live with 720p with an early availability of the game. The development cost (time and developers) of PS3 game was known to be really pricy.
With all these experience, it is not difficult to understand the direction that Sony has taken. However, I think it is very important for Sony to understand why Microsoft has chosen PowerPC on XBox360 instead of continuing the direction of using X86 as in XBox.
The game console folks have no loyalty to any specific architecture. Sony itself used MIPs, then moved to Cell and now x86. Xbox uses custom powerPC cores from IBM and may also move to x86 or ARM. Whoever gives them the lowest cost chip wins. In this instance AMD gave them the APU with both the CPU and more importantly the graphics processing.
While I understand that switching to a different architecture could happen on the game console, I am not sure if the cost of the CPU/GPU is the only reasons that motivate game console guys to do so.
Considering the headache of dealing with backward compatibilty issues of their game titles, I don't think this is a decision that they take it lightly.
Agree. IBM had designed the processors for all the 3 big game console providers and I suspect it was not able to provide a roadmap for them.
Reminds me of the decision by Apple to move away from IBM/Motorola to x86 for the same reason.
How much should they actually care about backwards compatibility?
Yes, they could encourage developers to port older PS3 games, or try to do a software emulation layer.
But the people who would want to play the old games already have a PS3. They might get a PS4, but will the PS3 simply be thrown out/sold/passed along, or will it be kept around for older games while the PS4 gets used for new content?
Backwards compatibility may not be that critical, if the PS4's performance on new stuff is compelling enough.
Agreed. In a game console, backward compatibility is actually a bad idea, as long as the developers support the new platform.
When people buy a new console, you want them to buy all new titles as well. The old games they already have were bought to be played on the old console.
That's an interesting and kind of important question. One of the problems Microsoft had with the original X-Box was PC compatibility. It was easy to put Linux on it. But Microsoft needed something like five title sales per X-Box just to break even on the hardware.
And given the actual computing demands of most people, the PS4 looks pretty able to replace a regular PC. Sony must have some plan to avoid the "wrong kind" of popularity.
Move away from IBM should be easily explained by IBM not installing enough wafer making capacity at 32nm and beyond. GF is IBM mfg partner. But they must have gotten a good deal from AMD , which also make chips at GF, using same tech as IBM.
It's the software stupid - something all companies big and small forget. They should have stayed with X86 from the beginning.
The PPC and CELL choices are made by hardware engineers that fall in love with speedy hardware specs, sell it to management and then throw it over the wall for the software community in the "other building"
I have to agree it is the software and to a lesser extent the hardware. If the games are ready and of high enough quality then who cares what the platform is? On the other hand if the hardware platform can't get out of it's own way then no amount of software can make up for the lack of responsiveness. I am guessing that the X86 tried and true platform with a lot of horsepower will win a lot of software developer support and therefor more/better games!
I wouldn't think it was a 32 nm capacity issue at IBM as I believe they have as much 32 nm capacity as they did at 45nm. I believe they still are supplying the newest game processors for Nintendo and Microsoft.
Seems like a well thought out path to follow. Given the relatively long product life cycle of Sony Play Stations. Employing AMD X86 architecture will be well suited for upgrades and enhancements, in particular GDDR6 has my attention, which should be going into production next year (if I remember right, AMD essentially created the GDDR memory standard for what it's worth). It would not be surprising if GDDR5 was skipped in the PS4 platform altogether. I find this news fascinating.
"the switch to the X86 CPU should also attract a large installed-base of “extreme PC gamers” to PS4, he added."
This was hilarious - as though PC gamers use a PC because of how much they love the x86 architecture.
Given there's about 0% chance PC games will run on the PS4, the choice of CPU is irrelevant.
The comment about PC gamers seems to ignore the why of PC gaming .... continuous hardware upgrades (its a hobby thing), cheaper games, sitting at a desk, versus in front of the TV environment, Skype while gaming, etc., etc.
I think the hardware is well targeted for the price point, but the average PC gamer will be well beyond this a few years into the PS4 life cycle.
I am not putting down the hardware, but if they think they are going to win back PC gamers, they are sadly mistaken.
Replay available now: A handful of emerging network technologies are competing to be the preferred wide-area connection for the Internet of Things. All claim lower costs and power use than cellular but none have wide deployment yet. Listen in as proponents of leading contenders make their case to be the metro or national IoT network of the future. Rick Merritt, EE Times Silicon Valley Bureau Chief, moderators this discussion. Join in and ask his guests questions.