Predict: The Next Generation Console Tech

[IllusionistK]

In hindsight, it seems Microsoft couldn't avoid spending a billion dollars on Xbox 360 be it on extra cooling/hardware or warranty claims. So the question is, how much extra hardware does a billion dollars buy you?

Is this a fair assumption or was it a cost to be avoided entirely?

If it's an unavoidable cost then doesn't that theoretically increase the thermal ceiling(by how much is anyone's guess) of their box? Either they a) put the billion dollars towards R&D or b) reserve a billion dollars for warranty claims.

I don't feel this is an irrational syllogism.

Thoughts?

 

[anexanhume]

In hindsight, it seems Microsoft couldn't avoid spending a billion dollars on Xbox 360 be it on extra cooling/hardware or warranty claims. So the question is, how much extra hardware does a billion dollars buy you?

Is this a fair assumption or was it a cost to be avoided entirely?

If it's an unavoidable cost then doesn't that theoretically increase the thermal ceiling(by how much is anyone's guess) of their box? Either they a) put the billion dollars towards R&D or b) reserve a billion dollars for warranty claims.

I don't feel this is an irrational syllogism.

Thoughts?

So you're ready to assert, that of the 50 million xbox 360's sold, it would have taken an extra billion in R&D and materials to not have to institute said program (200 dollars per console!)?

I don't buy it; engineering it right in the first place was the best solution, plus you have to consider the impact to their brand by the negative press. Planned failure with reparation costs built into the failure is not a solid business model. If you plan for your product to fail, it's because you expect the consumer to buy another one to replace it, not make a warranty claim.

 

[3dilettante]

In hindsight, it seems Microsoft couldn't avoid spending a billion dollars on Xbox 360 be it on extra cooling/hardware or warranty claims. So the question is, how much extra hardware does a billion dollars buy you?

The rumored actual fix was a re-engineering of the Xenos chip, or at least how it was affixed to its substrate. The alleged failure was one of thermal stress breaking the solder balls or chip layers that interfaced between the silicon and the substrate material.
A more robust cooling system could have reduced the failure rate or extended the lifespan by reducing thermal stress, but the eventual real fix was rolled into a die shrink. Either the new chip allowed a new packaging scheme, or its reduced power draw reduced the thermal stress.

The real case of false economy, if the rumors are correct, is that Microsoft tried to save money by delving into the physical implementation of a thermally active package at the time that solder and other materials were in a transition, an area it had less experience in than the chip companies like IBM and ATI.

If it's an unavoidable cost then doesn't that theoretically increase the thermal ceiling(by how much is anyone's guess) of their box? Either they a) put the billion dollars towards R&D or b) reserve a billion dollars for warranty claims.

The evaluation may be what they can model for mechanical and thermal stress on a complex IC and package. The more power it draws and the more its temperature varies, the more knock-on costs down the line to the actual chip, its packaging, cooling, and long-term survivability.

It doesn't take a billion dollars to either learn how to implement or contract for a quality package, or just avoid pushing the package into risky territory.

 

[Barso]

What I look forward to the most is which company will give us the more powerful system. Personally I think MS are in the stronger position to give us a higher-spec console but maybe they will think that there's no point any more.
Or do you think they will be more evenly matched than the Xbox360 and PS3 are now?

 

[jonabbey]

What I look forward to the most is which company will give us the more powerful system. Personally I think MS are in the stronger position to give us a higher-spec console but maybe they will think that there's no point any more.
Or do you think they will be more evenly matched than the Xbox360 and PS3 are now?

I expect whoever comes out later to have the genuinely more powerful system this time.

My vague understanding is that Sony was as late as they were with the PS3 because of issues with getting a reliable supply of 405nm blue laser diodes on tap, which caused them to launch when their GPU was very close to being past its sell date.

Edit: That may also (more) be because NVidia just didn't have anything better when it did launch. They were just putting out the extremely big and hot G80 which couldn't go into PS3 under any circumstances, so there was nothing they had with unified shaders.

It would have been great if Sony could have taken the extra year to put more RAM in the PS3 than the 360 had, but the Blu-Ray drive was already costing them a frack-ton.

 

[anexanhume]

The rumored actual fix was a re-engineering of the Xenos chip, or at least how it was affixed to its substrate. The alleged failure was one of thermal stress breaking the solder balls or chip layers that interfaced between the silicon and the substrate material.
A more robust cooling system could have reduced the failure rate or extended the lifespan by reducing thermal stress, but the eventual real fix was rolled into a die shrink. Either the new chip allowed a new packaging scheme, or its reduced power draw reduced the thermal stress.

The real case of false economy, if the rumors are correct, is that Microsoft tried to save money by delving into the physical implementation of a thermally active package at the time that solder and other materials were in a transition, an area it had less experience in than the chip companies like IBM and ATI.

My understanding was that this was because Microsoft wanted to own the actual design after having to buy parts from Intel for the first xbox. Hopefully they've learned and err on over-engineering this time around.

 

[Squilliam]

Just like they have no money to offer upgraded textures for the PC ports, right? There will be launch titles that look a bit behind, but within a year you'll have major AAA titles making last gen look fairly weak.

The thing with those upgraded textures is that they already exist in artist libraries. Aside from throwing otherwise next to no net cost upgrades onto the more powerful system they likely won't be throwing $60,000,000 at art assets to justify a next gen experience on a <20M userbase.

 

[AlphaWolf]

I never said they'd be throwing massive dollars at it just that they won't be developing for last gen hardware 2 years into the next product cycle.

 

[Rangers]

I expect whoever comes out later to have the genuinely more powerful system this time.

I dont, since I think each node gets more difficult and time consuming. Probably the opposite. One year might have been a big deal in 2005 in tech time, but I think not so much in 2013. Accordingly the advantage of launching first probably becomes greater since you will give up less power to do it by my theory.

It will really come down to each manufacturers "will" to be powerful imo, more than time frame. And heck how long has GDDR5 been state of the art? I dont even know if there's better on the horizon...

 

[TheWretched]

Also, launching early gives you the advantage of a lot of devs starting development on your system. That's probably one of the reasons why very early PS3 games were pretty bad ports (amongst others, obviously)

 

[JardeL]

What are the chances of they [Sony/MS] both going with AMD CPU&GPU ?..

 

[AlphaWolf]

AMD cpu's don't seem to be that good in perf/watt atm.

 

[JardeL]

But AMD CPU&GPU combo may be cheaper then IBM CPU&AMD GPU, wouldn't MS/Sony choose price/perf over perf/watt?..

 

[Prophecy2k]

My Q9550 and 6950 manage fine in my P182 with two low RPM 120mm exhaust fans ... that's not a lot of surface area, a console can manage that too. I don't see what the big deal is.

Surface area isn't the only factor. Two 120mm fans will move a considerable amount of air through the unit even at relatively low RPMs. The fact that you're cooling with a shoddy medium and even have a relatively small amount of surface area for heat exchange can be gotten around by transporting more of your cooling medium across that surface.

Anyways, my point with that post was a reply to the person who was asking about any revolutionary cooling technologies. I was essentially saying that you pretty much won't see any cheap enough for a console for the forseeable future that aren't just a simple variation on passing ambient air through the unit.

 

[Shifty Geezer]

Are there even any expensive revolutionary solutions that are viable in anything other than big-box PCs? AFAIK there's no progress in cooling because there's no other tech possible beyond changing medium (liquid cooling). All solutions will involve fans and heatsinks, for decades to come, without some radical new tech appearing (beaming heat energy directly into the household boiler!).

 

[Prophecy2k]

Two 12 cm fans would seem to require an awful lot of surface area for a console - that's four times the area that the 360's two fans took up and it's not like you could just mount one of them directly over an optical drive - and you certainly couldn't put one along the edge of the case. Inside a PC case there's a cavern for the air to flow through (even in a mATX case), inside a console the space is much more limited and how the air moves through the case is at least as important as the maximum area of fan(s) you could cover the outside of the case with.

A 2kW low profile fan heater can fit into a space not much bigger than the launch 360, but I don't know how many people would want to put one in their AV cabinet.

Indeed, this is the biggest factor for something like a console. You can't simply have a massive chamber of air inside your console because the console itself would be huge and that would drive up sipping costs etc.

The space that air has to flow through within the unit is directly related to the power required to drive the fans. If the space is large you ONLY NEED a low power low RPM fan because the pressure drop through the space is low. Constrict that space however, as in a console and you need higher power, higher RPM fans to provice enough pressure head to drive the air through to remove enough heat. You're limited also within a console by heat transfer area, as well as all your heat producing components being centralised at one point (unlike say a desktop where your CPU is on your motherboard and GPU on it's add-on board), so this makes your heat transfer less efficient anyway becuase you'll have a lower coefficient of heat transfer.

So a 200W console (total system power) and a 200W (total system power) desktop PC are an apples to oranges comparison in terms of designing a competant cooling solution that can remove enough heat, produce as little noise as possible (also important), and also not be prohibitively expensive enough to remain competitive against your competitor's console. Cooling in next-gen consoles will certainly not be a trivial thing.

 

[MfA]

Anyways, my point with that post was a reply to the person who was asking about any revolutionary cooling technologies. I was essentially saying that you pretty much won't see any cheap enough for a console for the forseeable future that aren't just a simple variation on passing ambient air through the unit.

My point is that we don't need anything revolutionary ... just more heatpipes. Hardly revolutionary and with enough of them the fact that air is a poor energy transfer medium is irrelevant given the potential airflow available to a console the size of last gen launch units.

As for restrictive ... the part of the XBOX360 not occupied by the drive section is a wasteland ... as long as they are build with a single PCB much larger than the drive section there will be plenty of empty space to allow in air from the sides and up the fans/sink. If you want to build a tight cube with everything stacked in the drive footprint ala Nintendo it gets harder, but meh.

 

[Prophecy2k]

Are there even any expensive revolutionary solutions that are viable in anything other than big-box PCs? AFAIK there's no progress in cooling because there's no other tech possible beyond changing medium (liquid cooling). All solutions will involve fans and heatsinks, for decades to come, without some radical new tech appearing (beaming heat energy directly into the household boiler!).

This is likely the only way i can see it happening. Basically sumerge all your console components completely in a liquid medium (basically a console board in a bottle). Then your system could either be connected to your home water mains or central heating system (so your central heating water pumps provides the required head to drive the water through your console).

Or you could have a completely convective system where the natural convection currents drive the fluid around a semi-closed loop (however you'd need some sort of cold sink). Your console components would run much hotter in this mode though, unless you used a different cooling medium, maybe some non-polar low boiling solvent, or low density refridgerant. Then you could desing a multi-phase system and take advantage of your medium's ability to capture considerably more heat during a phase change.

 

[Prophecy2k]

My point is that we don't need anything revolutionary ...

Well, i don't know. I agree but if we could, it would be nice so that we could go balls to the walls with TDP and not have to worry about cooling ;-)

just more heatpipes. Hardly revolutionary and with enough of them the fact that air is a poor energy transfer medium is irrelevant given the potential airflow available to a console the size of last gen launch units.

Heat pipes are made from copper or aluminium and thus are not inexpensive materials. They also only really spread heat within the console itself, removing heat through phase change of the internal fluid from the heat generating body to a heat sink. That heat still needs to be removed at the other end and that's air's job. Regardless of how many heat pipes you have you still need to get the heat from the heat sink outside of the console unit itself and you only have fans driving your air to do that.

As for restrictive ... the part of the XBOX360 not occupied by the drive section is a wasteland ... as long as they are build with a single PCB much larger than the drive section there will be plenty of empty space to allow in air from the sides and up the fans/sink. If you want to build a tight cube with everything stacked in the drive footprint ala Nintendo it gets harder, but meh.

The early 360s and PS3s still have probably the highest failure rates of any consoles to date, so that definintely tells you something about how easy or hard it is to cool them with what are relatively small fans through pretty tiny chamber spaces. They can keep the TDP's of this gens consoles at launch or go higher, and even design them in the same or similar sized boxes, but i tell you if they do that we'll either have consoles that fail HARD again, or ones that sound like flipping jet engines.

 

[Shifty Geezer]

The early 360s and PS3s still have probably the highest failure rates of any consoles to date, so that definintely tells you something about how easy or hard it is to cool them with what are relatively small fans through pretty tiny chamber spaces.

The 360 faults appear to be heat related. The PS3's however aren't because the chips blew from overheating, but because the solder cracked. PS3's fan has several settings to keep the chips under critical temperature, and it's rare for the fan to spin up to even moderate speed in normal conditions, so clearly the cooling solution was adequate to deal with the heat of Cell and RSX. The same heat output next gen could be dealt with in exactly the same fashion, producing a quiet system.