Xbox One (Durango) Technical hardware investigation

Final nail for the yield issues rumor!

In Citi conference call, MS stated that multiple millions of units will be available before the first of the year. So at least 2 million units before 14.

 

Yep.. they have known for a while most likely and have held it close

 

It´s what i thought too, also Penello himself told something in the line of not giving all the specs at that stage for competitive reasons

 

This hidden gpu is not true. Please stop!

 

The extra 2 APU's and 3 TFLOPS are hidden inside the external power brick. Notice the thicker than usual power cabling.

 

Who thinks twice? The mods? It's remarkable how many people seem to misunderstand entirely why they got their infractions, and then go complaining in the middle of some discussion. People posting unsubstantiated rumours as facts in a discussion not about the rumours during silly season can expect to be reprimanded for being OT. Doesn't matter who your source is, unless you a proven developer working on HW known to be telling the truth, you can't post rumour as fact in discussions because of how that affects the discussion.

That eastmen's upclock rumour turned out be true doesn't justify him posting rumour as fact, nor does it give him carte blanche to post his sources as fact in general discussion. How exactly is one supposed to know which posters are allowed to post their rumours as facts such that the mods can know when to think twice about removing rumours from a discussion outside of the rumour discussion deliberately created to manage that side of discussion?

This is of course OT, but if I removed the OT remarks I'd be accused of censorship and silencing the dissenters.

 

I am more interested about these 15 co processors,what they can do and if they offload the CPU and GPU,what kind of graphics task they handle or flops performance.

 

May the dGPU rumor be either true or false but why MS uses high quality inductors that can feed the SoC up to 250w when the Soc itself is targeted for 100w TDP?

The SoC is labelled by MS in the slides released during Hotchips conf as "Main SoC". Doesn't this suggest something else?

 

Maybe, if those were the only two pieces of evidence. But next to everything else, interpretation is a lot less exciting. "Main SoC" then reads "Main component, the SoC" instead of "Primary SoC, the Secondary SoC is elsewhere which we haven't labelled."

 

As already noted, there may be multiple reasons.
1) There is some safety margin built in.
2) The board that was pictured may have had extra power delivery so that the chip's balance of clocks and power between the GPU and CPU can be tested without the board's power delivery getting in the way.
3) The latest APUs allow the CPU side and GPU side to ramp up and take most of the TDP all on their own if the other half doesn't need it. If the GPU and CPU have separate feeds, how can they ramp up if the board won't give them the power?

 

Have you ever read this patent by MS? It reminds me the old rumor about Yukon architecture...

http://appft.uspto.gov/netacgi/nph-...090.PGNR.&OS=DN/20120159090&RS=DN/20120159090

 

If you search the forum for that number, you will find where it's been mentioned before.

 

I remember at Hotchips or the Architecture Panel the MS architects talking about how there was contingency built into the system and the design. That if silicon/processors fails that there would possibly be others to take that burden..

Is it possibly for some of these custom co-processors/dpus etc. to possibly do this?! Such that maybe some of these 15 processors other than CPU/GPU could do different roles if they needed to? thus the power being able to ramp up/down when needed?

I'm not a HW guy BUT ive heard some say that eg. the Audio processors were equivalent to 8core jags and there dpu nature could mean they could possibly do other tasks if needed (reprogrammed) ?!

Is this even remotely possibly? Is this the contingency they were talking about?!

p.s. these are the tensilica DPU's that the Audio processors are based on - http://www.tensilica.com/learning-center/dataplane-design

 

Its been mentioned many times before, and most of the applicants on that patent are the architects publicly speaking about Xbox One.. and one of those applicants, Mark Grossman is a partner architect for Microsoft that does work with AMD/ATI and is listed in a couple of ATI patents..

Of course none of this guarantees that the Xbox One is that design... just all speculation

p.s. don't read too much into that patent especially the diagrams that clearly show multiple gpu's ... that argument has been made many times before

 

There are multiple small processors in the various media blocks and specialized parts of the chip.
Power management has its own microcontroller or two, although if it's similar to current schemes it wouldn't be an offload processor because it should operate autonomously.

I don't know what quotes mentioned contingencies in the architecture.

 

No. They are custom processors. You can't program a hardware video encoder to sort a set of data structures, or a DMA unit to filter an audio sample.

Contingency would mean redundant components on the die, so a fault can be endured. This is standard practice for ICs. eg.
PS3's had 8 SPEs, but one of them was redundant so that when 1 SPE didn't work, the CPU as a whole still had 7 SPEs to use. You can put in an extra CU or two on a GPU and produce a final chip that still meets requirements even if a fault renders a CU useless. If the chip is 100% functional, you deliberately disable components to match the specifications, otherwise you'd have consoles with different processing power.

 

I could be miss interpreting this, highly likely since my field is software development, but here's the quote from John Sell at Hotchips

ref to quote: http://www.itworld.com/hardware/371478/microsoft-emerges-cpu-force-xbox-one-and-kinect

 

The RROD wasn't due to silicon defects, so unless Microsoft specifically mentioned it, I would assume the latter claim is the author's own misinterpretation.
The redundancies, as mentioned before, are extra units that are present in silicon in the event a manufacturing defect renders one or more of them inoperable. It doesn't cover post-manufacturing failure.

 

There's the main SoC and the Camera SoC from their slides (which I assume resides in kinect itself)... Probably it's just that...

 

I'm not sure where complexity figures into it... In the demo, the person on the tablet was purported to be in a remote location. Perhaps commuting home on a bus. You don't pipe graphics over a 3G connection. You just don't. Even piping it out over a typical home upsteam network connection is a little unbelievable.

The least-crazy explanation is that the tablet is running a little "The Division Lite" app, and it is drop-in connecting to the same network match as the other players. (An interesting concept. I hope it actually happens.)

[EDIT] Sorry, I didn't realize I was so many pages behind in this thread.

I'll add a question: I saw the term "diffused" printed on the XB1 SOC. Someone in this thread also used it. What does it mean? Is it a fabrication step?