Predict: The Next Generation Console Tech

[AlNom]

Aren't yields also a function of voltage as well as clockspeed? It's not just a question of how many chips simply work but also how many work at the set clockspeed and thermal parameters?

That's what I meant from mentioning the following:

it's not just whether the chip works or not, but also getting chips at the desired clockspeeds (or rather, chips that behave within operational power and thermal designs)

Power consumption can behave differently at the same voltage depending on, for example, gate design, the doping or just a synergy of random defects that make the chip crappier than others.

So if worse comes to the worse they could go with say a vapour chamber cooler initially and then spec it down to a regular heatsink/fan combo when yields improve and required voltages drop?

You mean on the same node? Potentially, but they'd probably want to play it safe and easy with just a single production configuration until a major revision. They'd also be able to skip testing chips as though they were binning them when that's not an actual consideration. Probability of chip failure is increased with higher temperatures, so they might as well just stick to the one type of heatsink that covers the worst case (IIRC, MS's TRC for randomly sampled units was 36 hours operating under load before failure).

One burning question I have is whether they could possibly get a console out in 2011 as those rumours suggest or early 2012 even on 28nm since we now have two contract fab companies slated to offer 28nm production this year? Thats the real question since it would have an extreme bearing on what they could pack in at their desired thermal limit. Although in saying that 32nm SOI is also a possible candidate although I don't know how much capacity GF can spare to anyone who isn't called AMD!

Might be too supply constrained. IIRC, Global Foundries isn't really well positioned for mass production until 2012.

Plus we still haven't even seen desktop graphics cards on 28nm from AMD/nVidia, and these two have lower scale production i.e. they've never required 1-2 million working chips at launch.

Anyway, which node do you think?

One would love smaller than 28nm, but that should be quite mature by 2013.

 

[Squilliam]

You mean on the same node? Potentially, but they'd probably want to play it safe and easy with just a single production configuration until a major revision. They'd also be able to skip testing chips as though they were binning them when that's not an actual consideration. Probability of chip failure is increased with higher temperatures, so they might as well just stick to the one type of heatsink that covers the worst case (IIRC, MS's TRC for randomly sampled units was 36 hours operating under load before failure).

What I mean is for them to go with a less than ideal configuration with a better than ideal cooling solution to start with and then transition from that configuration as the process improves and the required power for the same number of good candidates falls, perhaps after a respin some time in the future the same way that Nvidia was able to improve their chips considerably in a 12 month period through respins and process improvements. So after say 12 months a silent revision comes out which changes the heat sink to a general purpose design.

Might be too supply constrained. IIRC, Global Foundries isn't really well positioned for mass production until 2012.

Plus we still haven't even seen desktop graphics cards on 28nm from AMD/nVidia, and these two have lower scale production i.e. they've never required 1-2 million working chips at launch.

Surely if we were talking about GPUs or CPUs for the mass market they would infact need 10M+ of them every quarter whereas with a console all likelihood is that they'll need 4M or fewer by the time they ship their first console. So for instance they may not be able to launch Bobcat/Ontario replacements on 28nm in 2011 but they might be able to swing the volume for a console. I don't know, im just speculating... They should be able to yield decent numbers of good chips per wafer if the design itself is less than 140mm^2 all up, knowing Nintendo this is likely to be the case regardless. If we were talking about 200mm^2+ Xbox next or PS4 chips it might be a different story however.

 

[Nesh]

I was reading the other thread about using Cell/CPU architectures as a GPU and it kept me thinking.

Its generally accepted that a CPU architecture is not as efficient or as capable as a GPU in doing what the GPU does as a dedicated processor for graphics.
But, the Cell has proven that CPU architectures although may be bad in doing graphics as good as GPU's, some may be perfect in assisting the GPU significantly.
First party studios did amazing things with the SPU's.
So what if lets say Sony except from the normal CPU, also includes a Cell derivative in the next Playstation dedicated specifically to work with the GPU?
It could be a cheap way in giving an extra flexibility and performance for the developers to experiement with.

 

[stiftl]

CPU wise, probably barely faster or even slower. There was a talk some page ago about how Bobcat doesn't even competitive with XCPU in single-thread performance.. Krishna should have an enhanced version of Bobcat, but we don't expect miracles.
On the GPU side (and rough power wise), Ontario/Zacate hit the 80 Gigaflops mark. Xenos is around 192 Gigaflops. Krishna GPU could be more powerful than Xenos, giving the more advance production process, the DirectX11 features, update technology etc.

In the end, Krishna should be close or a little "faster" than the current Xbox (not taking in consideration RAM and bandwidth).

It wouldn't solve one of the biggest problem of Nintendo Wii: the lack of good third party support.

Trinity seems a better choice for Nintendo imho. It should be considerably faster than Xbox360/PS3 in both CPU side and GPU side, and probably it won't be that much slower than Xbox 720 / PS4 (if this 2 goes for CPU + GPU die). This would allow for good conversion of third party games.

Sure it would be a better choice, but unfortunately Krishna will be a lot cheaper i guess. And as we know Nintendo I think we shouldn't write it off because of that.

Otherwise I would think that with Trinity nobody of the massmarket would see a difference later on PS4/XB720 and this could be the chance for Nintendo to clear a lot of ground.

 

[DeadlyNinja]

You guys forgot to link to this post.

http://www.neogaf.com/forum/showpost.php?p=26403963&postcount=464

Chances are, it's Microsoft that's readying a new console from the looks of things.

 

[rpg.314]

Changing the physical design of the product and the electrical design of the device involve re-designing and then validating it.

Does a silicon respin require re-validation? I should think not.

 

[AlNom]

You guys forgot to link to this post.

You forgot to read the last page.

 

[Barso]

If this means that MS are going with Krishna in their next console then does this make the xbox360 successor a slight upgrade or a next gen technological marvel?

 

[AlphaWolf]

You guys forgot to link to this post.

http://www.neogaf.com/forum/showpost.php?p=26403963&postcount=464

Chances are, it's Microsoft that's readying a new console from the looks of things.

and how exactly does that post convince you of that?

If this means that MS are going with Krishna in their next console then does this make the xbox360 successor a slight upgrade or a next gen technological marvel?

Krishna won't be chasing the high end. It could easily still be plenty more powerful than current consoles. I don't expect them to set the target so low. Nintendo however...

 

[Squilliam]

I don't see where that post specifically points to a new Xbox console coming out at the end of 2012. Its a lot of supporting evidence towards that end but in the end it doesn't really give us much to work with aside from food for thought. I can't even say much on which CPU they would implement there, probably Bulldozer IMO due to higher single thread performance but then again if the enhanced Bobcat gets say 15% per clock and clocks 50% higher it could be a decent contender.

 

[liolio]

We could probably use the info we have on the 360 design & manufacturing timeline to get an idea of how rapid things can be.

So sure, two years would be an insanely huge amount of time if the design is done. If it's a cutting edge process, there will be problems with getting good yields - it's not just whether the chip works or not, but also getting chips at the desired clockspeeds (or rather, chips that behave within operational power and thermal designs). It'll take some time to experiment/optimize the chemical process as the console companies can't rely on binning as Intel and AMD do. Still I wouldn't expect that to take too long to fix with that sort of timeline (much less even 6months to a year) as they can focus on fixing these sorts of production issues instead of trying to push out chips as fast as possible.

On a more mature process, the problems should be pretty insignificant unless we're talking monolithic 400+ mm^2 chips. Then it's just a matter of ramping up production.

May be Ms acts the same way as Nintendo claims to ie they are constantly evaluating hardware and I don't which higher ops in Nintendo told that, if needed they stop evaluation and move to production in which case two year may be enough.
ANyway rumours are rumours so there is not much to argue about but I can't dismiss Ms launching fall 2012 there is another argument in favor of Ms doing so the fall 2012 seems like the good time line for Nintendo to launch they my not want to let them the benefit of an early launch and I'm not sure that they would pass on the possibility of getting Sony its pant down again.
I hope we will get more solid informations soon.

 

[Squilliam]

May be Ms acts the same way as Nintendo claims to ie they are constantly evaluating hardware and I don't which higher ops in Nintendo told that, if needed they stop evaluation and move to production in which case two year may be enough.

Whilst Microsoft may be competitive with Sony and Nintendo directly the Xbox 360 comes from the EDD which is Microsoft's version of Apple. So I wouldn't be surprised if they aim to use Apple-esque product strategies in order to compete in the game business. In order to be competitive with Apple they'll need multiple hardware teams working on projects years in advance of release. This has been their weakness IMO but at the same time from the current and past hiring sprees they seem to be building that into a strength.

What I figure could be their MO going forwards is to create a product which gets the basic things right and then continually expand upon their offerings in order to increase revenue and competitiveness. I can see them going for strict TOCs (right word?) for developers to follow and going for annual, bi-annual, or tri-annual tick tock hardware cycles. They have mentioned that they want more fluid console cycles in the past and X86 may be the ticket to do just that.

Lets see a possible timeline:

2012: Xbox next releases.
2013: Nothing? PS4? Releases?
2014: Xbox next refresh.
2015: Nothing?
2016: Xbox next next releases?

If they keep within a strict architecture such as X86 and development environments like DX11+ they could easily push out revisions to hardware if and when desired or needed. Furthermore I could also see their saving big software changes for alternative years. The years where I wrote nothing could actually represent big Live feature implementations.

Overall they would see greater flexibility if they release on this kind of production schedule. They won't have to worry about whether 3D takes off in 2014 because by that time they could have a refresh out which can more than cover the required performance and at little additional cost. That way they won't have to pack in hardware in 2012 in order to be relevant in 2014 and even if Sony does do that then they can come right back out with a refresh which is more than competitive the following year. Beyond this they'll likely make back additional hardware expenditure with additional hardware revenue as Apple has discovered with a set percentage of last years Apple adopters also aquiring next years product with the added bonus of being able to appeal to people they may not have appealed to with the original hardware.

It'd be interesting to see if they have another hardware revision lined up for this year on the Xbox 360, their actions in respect to the legacy 360 may have significant bearing on predicting their future.

 

[(((interference)))]

http://www.computerandvideogames.com/292541/news/just-cause-studio-prepared-for-xbox-720-arrival/

Sundberg said that "everyone is speculating" over the timing and technological make-up of both Sony and Microsoft's next console, but opined that "everyone agrees" on one thing: - There will be no optical disc compatibility.

I certainly hope not! What would I do with all my 360 games? I really do hate digital content, can't resell it, can't lend it to friends etc. and if you have slow net speeds who wants to wait a couple of hours while the latest game you've been after downloads.

Hopefully they'll learn from the disaster that was the PSP Go and stick with physical formats.

 

[AlphaWolf]

I certainly hope not! What would I do with all my 360 games?

Play them on your 360?

 

[Shifty Geezer]

I certainly hope not!

Reading it in context, I think he means no compatibility with XB360's optical disc games. The quotes are fractured and this line isn't even a direct quote! "No optical disc compatibility" would be a weird way to say "no optical drive" or "DD only", whereas it is an suitable, if awkward, way to say no compatibility with 360's disk titles but could be compatible with download titles, assuming download titles are fully XNA compliant where disk titles are more adventurous in use of the hardware.

 

[Nesh]

http://www.computerandvideogames.com/292541/news/just-cause-studio-prepared-for-xbox-720-arrival/



I certainly hope not! What would I do with all my 360 games? I really do hate digital content, can't resell it, can't lend it to friends etc. and if you have slow net speeds who wants to wait a couple of hours while the latest game you've been after downloads.

Hopefully they'll learn from the disaster that was the PSP Go and stick with physical formats.

I dont think this is practical.

1) Both will be removing Blu Ray+DVD playback for movies which defeats the purpose of an all in one media system. Also its like Sony is removing one of the features they promoted so much as a competitive advantage on the PS3

2) Not all countries have high speed connections to rely on digital distribution

3) They will require very large hard drives which will be filled fast

4) Retailers wont like this. Consoles have low profit margin for them. Almost zero. They want to sell games

 

[msia2k75]

Maybe they meant "optimal" instead of "optical"?

 

[Prophecy2k]

Reading it in context, I think he means no compatibility with XB360's optical disc games. The quotes are fractured and this line isn't even a direct quote! "No optical disc compatibility" would be a weird way to say "no optical drive" or "DD only", whereas it is an suitable, if awkward, way to say no compatibility with 360's disk titles but could be compatible with download titles, assuming download titles are fully XNA compliant where disk titles are more adventurous in use of the hardware.

I'm in Shifty's boat here. I think it's a reference more towards BC for disk-based games rather than no optical drive altogether (which would equal suicide).

It's either a mistranslation or utter BS. No in between

 

[3dilettante]

Does a silicon respin require re-validation? I should think not.

In the scenario I responded to, it would be everything else around the chip that would need validation.
The heatsink and electrical design would be changed for what amounts to a respin of the silicon, and those changes would need additional validation.

 

[Scott_Arm]

Reading it in context, I think he means no compatibility with XB360's optical disc games. The quotes are fractured and this line isn't even a direct quote! "No optical disc compatibility" would be a weird way to say "no optical drive" or "DD only", whereas it is an suitable, if awkward, way to say no compatibility with 360's disk titles but could be compatible with download titles, assuming download titles are fully XNA compliant where disk titles are more adventurous in use of the hardware.

I have to agree on this. I'm sure both will have BluRay drives for movies. And it would suck for Canada because of the impending bandwidth cap doom. They can' t build a business strategy assuming everyone is on really high-speed broadband with huge caps.