Best news:
Hopefully not just empty words.
"We have always started with the core audience and then expanded. A console always needs a solid core of games that appeal to gamers. Look at God of War. We launched that in the seventh year of the Playstation 2 and a lot of people wondered why we did. It's because we always wanted to keep the support of the core gaming audience.
That's not to say we're not doing anything to expand the demographics. Singstar and Buzz are obvious examples. But we need to do this in a controlled way. If you go mainstream too quickly and don't support the core gaming audience then you lack the pillar to support your platform. Without this pillar you end up with a fickle audience that might be big but will probably move on."
Best news:
Hopefully not just empty words.
While I'm all for grand technology and innovation, and I believe the PS3 is a more powerful machine than the 360, I just feel Sony has their priorities all mixed up, and they rarely seem to learn from their past mistakes.
With this current generation Sony saw their market share drop off dramatically. Microsoft showed with the 360 that simply have a competent machine technology wise, was enough. It shared much with current PC's, which made game development quite simple and straight forward. Graphically speaking it is capable of things very similar to what the PS3 can do.
So what does Sony seem to be doing for it's next consol.e? Something very similar to what they did with the PS3. Custom chips, shooting for the moon, etc, etc.
Why? It seems so much more prudent to wait to see Microsoft's specs for the next console, and then beef yours up by a touch if you want to say you're more powerful. If the machine was actually easy to program for it would be even more likely that developers could take advantage of that extra power.
But no, Sony will do things it's own way, have something that is probably significantly different from the next Xbox chip wise, not make online features the focus it needs to be, and ultimately lose more money and market share.
I'm firmly convinced this company is capable of driving itself right off the financial cliff in the next 5 years.
Soc at 28nm doesn't bode well for performance. Maybe it's for something else.
I'm just scared that they change to the way Microsoft changed to in 2009 where basically the only AAA games that were aimed at me that was published by them was Halo: Reach and Gears of War 3.Unlikely.
They have been able to successfully publish core and casual games without any problems, I don't see why that would change in the near future.
Yes, the obvious path Sony would take would not be dumbing down their aspirations for hardware, but to make more efforts for development on that hardware to be easier for first and third parties. I think we saw plenty of examples this gen of developers closer to Sony being able to prove that the PS3 was the most technically advanced console of the generation. However, that doesn't matter when third parties are struggling to produce multiplatform games on par with the 360 versions. If SCE is able to provide more resources to make developing for their unique hardware easier, every game could potentially outshine all other versions and that would actually be noticeable.
Technical prowess is less important than doing something interesting. Squeezing all they wanted into the PS3 hurt them, as did launching late. If they release yet another $600 system a year late it's probably not going to matter what their dev tools are and whether it supports 4K.
Except they have been doing more and more to ease development.
Their developers share their tech (with 1st and 3rd parties) and they even have their own multiplatform engine.
On top of that, if the Vita is anything to go by, they are taking more and more developer's advice when it comes to console development.
- 'the company is working on a system-on-chip (SoC) to underpin the product for "seven to 10 years".'
Very interesting article because we're not talking about rumors.
So basically PS4 will be based on a SOC at 28nm. Rumors suggested the same thing for Xbox next.
And as I said back then the SOC solution means that we won't get state of the art performance in the next consoles. It can still be very powerful but it's nowhere near the power of having two saperate chips at the same die area.
why use 4D ? 3D allready failed. I got even a 3D tv and never use the 3D function..
4D 300K 8K cat videos, my body is ready. I doubt that even PS4 will go with 4K route but I guess it's possible considering they are already doing it with PS3 to some extend.
Don't know if I'd read anything definitive into the 'system on a chip' business just yet.
He talks elsewhere about "an architecture where the bulk of processing will still sit on the main board, with CPU and graphics added to by more digital signal processing and some configurable logic.” - which sort of suggests a more traditional discrete approach for those components.
Let's remember that Cell was referred to as a 'system on a chip' design too...
That said, a design which brought the CPU and GPU onto one die for reasons of latency could indeed be on the cards. But I think we need a bit more to go on.
If he's looking around at work on stacked chips, a lot of it currently may be coming in 28nm flavours. It may not necessarily represent their target process... data re. yields on their target may not really be even available yet.
But discrete stacked components is far better than SOC when it comes to performance. The only benefits of SOC is die size and cost reduction. Hopefully Sony will devote 500 m^2 to the GPU and CPU again. SOC only allows for 300m^2.
Bullshit, you fuckers. PS3 was supposed to be 120fps and 1080p, amirite? Maybe when I have kids.
DSPs as in the SPEs?
This is all assuming Sony is still around in 2020.
SPEs are DSPs in a sense.
If we go by the SOC theory at this point I wouldn't be suprised if indeed the CPU was an optimized Cell with 32 SPEs and 2 to 4 PPUs.
That's what the PS3 was supposed to have at one pointiamshadowlark said:32 is a bit high though right?
IIRC, before E3 05, Kutaragi did a presentation where he talked about realtime data off the network as being a next step for games - that it would introduce time, 'the 4th dimension' to gaming. E.g. in a racing game, getting up-to-the-minute race results in-game, realtime weather data from the network feeding into track conditions etc. That's all he meant, but the Krazy Ken meme jumped on it
This sounds like a bunch of gobbledegook and most of it not remotely feasible anytime soon. Very scattershot and no clear vision either.
Has Sony not changed at all since Kutaragi? I can only assume it's a bad translation and, most of what is written wasn't meant in the context it's implied in.
Otherwise this thing will take so long to come out it will be competing with Xbox 5. 8kX4k at 300 FPS? I mean, really.
That's what the PS3 was supposed to have at one point
32 is a bit high though right? I know Rambus has this TB bandwidth and all but damn.
This sounds like a bunch of gobbledegook and most of it not remotely feasible anytime soon. Very scattershot and no clear vision either.
Has Sony not changed at all since Kutaragi? I can only assume it's a bad translation and, most of what is written wasn't meant in the context it's implied in.
Otherwise this thing will take so long to come out it will be competing with Xbox 5. 8kX4k at 300 FPS? I mean, really.
Not really that high, if you think that SPEs are somewhat similar to shaders and we have hundreds of those in modern GPUs, Kutaragi originally wanted 4PPUs and 32SPEs in the PS3 and IBM thought he was crazy LOL LOL
Btw effectively having the 32SPEs, 2to4 PPUs Cell as the CPU part of a 28nm SOC would ake a lot of sense, it's small enough, it's good jump from the PS3 CPU in terms of performance (in the 5X-10X range), there is little R&D, it keep backwards compatibility with no added costs and it allows for a quick learning curve for developers.
We'll see though...
Almost everything he's aiming for with this tech is something I either couldn't care less about, or would actively dislike. Enough with the gimmick race already
Not sure what segment of the "core" is interested in yet another round of novelties that wear off after the first year or two. Can't remember the last time I played a game that even used Sixaxis, let alone used it well. Couldn't care less about the Move, if I wanted that crap I would have bought a Wii. And now he's aiming for, let's see, augmented reality? Lame. Haptics? No thanks, I've used the Novint Falcon. Another axis of accelerometer? I'm sure developers will just have a field day with that, can't wait to download the one $5 game that uses it well and then forget it exists. Measuring my vital signs? Get out of here with that crap, who's asking for it? Aiming for 8K/4K? Yeah that's exactly what we need after struggling so hard with even 1080p.They still seem to be aiming for the core.
Not sure what segment of the "core" is interested in yet another round of novelties that wear off after the first year or two. Can't remember the last time I played a game that even used Sixaxis, let alone used it well. Couldn't care less about the Move, if I wanted that crap I would have bought a Wii. And now he's aiming for, let's see, augmented reality? Lame. Haptics? No thanks, I've used the Novint Falcon. Another axis of accelerometer? I'm sure developers will just have a field day with that, can't wait to download the one $5 game that uses it well and then forget it exists. Measuring my vital signs? Get out of here with that crap, who's asking for it? Aiming for 8K/4K? Yeah that's exactly what we need after struggling so hard with even 1080p.
Stuff like a good SDK and an improved architecture? Now we're talking. I really wish they would just refine the technology that already exists and even prune away some of the technology that isn't well-used. Even if it's just for one generation. Just chill out with trying to do wild and wacky new ideas, turn around and put as much effort into perfecting what we've already got, maybe even introduce a new thing here or there if it's genuinely relevant and useful, and then go back to pushing forward if you have to
I remember 4D demos with real time ageing of surfaces. A bathroom going from spotless to decrepit. That sort of stuff.
Not really that high, if you think that SPEs are somewhat similar to shaders and we have hundreds of those in modern GPUs, Kutaragi originally wanted 4PPUs and 32SPEs in the PS3 and IBM thought he was crazy LOL LOL
Btw effectively having the 32SPEs, 2to4 PPUs Cell as the CPU part of a 28nm SOC would ake a lot of sense, it's small enough, it's good jump from the PS3 CPU in terms of performance (in the 5X-10X range), there is little R&D, it keep backwards compatibility with no added costs and it allows for a quick learning curve for developers.
We'll see though...