Xbox One (Durango) Technical hardware investigation

Status
Not open for further replies.
I think the counter argument is there are many GCN parts reliably clocked above 880Mhz, The question is how much thermal increase their box can handle.
If you're designing a mass-produced CE device, you need to cost in a failure rate. If an 800 MHz part will net you a 2% failure rate, and 880 MHz a 12 % failure rate, even though the 10% increase is 'doable', it'll come at significant cost. these clock speeds are selected for very good reasons. ;) Clearly Sony found that the parts they were getting for RSX just weren't reliable enough at 550 MHz so had to downclock, which shows how sensitive the issue can be. (ironically PS3s still died in droves, so maybe they shouldn't have bothered? :p)
 
I also don't see ppl diving into discussion about how the eSRAM, DME's and RAM in Durango work as a whole either unfortunately.

Yeah, I would like to see this talk about more. I have no ideal what the general consenus here around this particular aspect of Durango.

Is DME basically a fpga configured as a DMA for a gpu? How old is CudaDMA, which seems to be a software based solution?

http://research.microsoft.com/en-us/projects/fpga_apps/

This paper was released by MS just last August.
http://research.microsoft.com/pubs/172728/20120628 Speedy PCIe FPL Final.pdf

This paper discusses difficulties and insights related to the implementation of the PCIe protocol on the PC platform in the form of the Speedy PCIe core and offers it as a solution or starting point for future research. The Speedy PCIe core delivers a general purpose solution that solves the problems of high speed Direct Memory Access (DMA) while offering an interface that is generic and adaptable for a large number of applications.
 
If a 10% overclock is easy to pull off, how come RSX got reduced from 550 MHz to 500? Why was Xenos 500 MHz and not bumped up to 550?

Weren't those machines already pushing the limits of console-sized boxes with cutting edge technology though? These machines seem much more watt, heat, noise friendly to begin with and pushing whats inside by 10% still doesn't bring them close to what the hot-rod consoles of yore were doing.

In relative terms its the difference between an Indy car going 10% faster from 220 to 242 mph (insane), and my sedan going from 70mph to 77mph (largely inconsequential).

I think you made an even better point around all the custom knick-knacks in here and accounting for all their various tolerances makes it more difficult than a more commonplace CPU/GPU setup to be "OC'd."
 
If you're designing a mass-produced CE device, you need to cost in a failure rate. If an 800 MHz part will net you a 2% failure rate, and 880 MHz a 12 % failure rate, even though the 10% increase is 'doable', it'll come at significant cost. these clock speeds are selected for very good reasons. ;) Clearly Sony found that the parts they were getting for RSX just weren't reliable enough at 550 MHz so had to downclock, which shows how sensitive the issue can be. (ironically PS3s still died in droves, so maybe they shouldn't have bothered? :p)

Sure there's going to be associated costs and any decision is going to be risk vs reward just like adding ram. I wouldn't be surprised if they were an unknown at MS at this point and they are just targeting AMD's lowest bin.
 
That's not true. PS3's cooling solution was superb, well implemented (maybe not the thermal pad), and certainly Sony's engineers aren't stupid enough to fail to go through their thermal management maths and cope with the heat. PS3's fan had several speeds - if it was overheating that caused the issues, the fan would spend time on the fastest, very loud, setting, before the console went belly up and the chips burnt out. You should also have seen overheating errors like corrupted visuals before the final death, similar to overclocked, burnt out PCs. That wasn't the way with YLOD. PS3 just wouldn't switch on, caused by fatigue in the solder points, proven by temporary heat gun fixes that wouldn't do jack if the component had fried. We also have independent repair companies telling us it was the solder.

MS's situation is maybe more debatable, but had the solder been leaded, there's reasonable chance MS's hot chips would have survived just fine even with board buckling.

The YLOD issue became a catch-all term to describe more than simply the failure case you described. My EU 60GB launch unit actually did overheat. The chips did fry and I did get the fan spinning up to its loadest (sounded like a jet engine) and then displaying wierd visual artefacts on screen before the PS3 OS actually kicked me out of the game I was playing and told me it was going to switch itself off becuase it had overheated. The issue I had was certainly overheating with my PS3.

In any case I was referring moreso to the RROD issue on the Xbox with my first paragraph, especially since that was the most prolific issue and was indeed a symptom of more than just a switch in the solder used.

Also, we only had one independant repair company in that Eurogamer article speculating as to what the issue was. The company that reflowed my PS3 when it died told me something different, of course based on their own experience with the consoles. My PS3 still failed again when I got it back from those guys too. Basically the chips were knackered.
 
If you're designing a mass-produced CE device, you need to cost in a failure rate. If an 800 MHz part will net you a 2% failure rate, and 880 MHz a 12 % failure rate, even though the 10% increase is 'doable', it'll come at significant cost. these clock speeds are selected for very good reasons. ;) Clearly Sony found that the parts they were getting for RSX just weren't reliable enough at 550 MHz so had to downclock, which shows how sensitive the issue can be. (ironically PS3s still died in droves, so maybe they shouldn't have bothered? :p)

Following your logic their AMD7770(1000Mhz,10 instead of 12 CUs) is sold with 12+% failure rate?:) Come on, the leaked Durango specs are not limited by its TDP(which I would guess around 100..120Watt). They should know better where the sweet spot between higher clock, more CUs and the TDP is. Last generation's first versions were around 200Watt with no sophisticated power management.
 
That's more because of market segmentation than any limit I'd bet. A10-5800k retails for $120.

Maybe you're right though and MS is pushing the thermal limits of an APU and Sony is screwed.

Yeah I was only talking about the overclocking the gpu as part of of an apu. Increasing the clocks 10% would probably have little effect on the gpu by itself but I worry how the increased heat might effect the rest.

Btw, there's the other risk with overclocking in the 10-15% range, if there's no issue to doing it then Sony can match you.
 
Yeah I was only talking about the overclocking the gpu as part of of an apu. Increasing the clocks 10% would probably have little effect on the gpu by itself but I worry how the increased heat might effect the rest.

Btw, there's the other risk with overclocking in the 10-15% range, if there's no issue to doing it then Sony can match you.

It depends when that final spec becomes known. Sony's apu will already have higher thermals (6xcu > esram), but yes that then becomes a value proposition for them, but I doubt Sony would react to MS closing the gap a few %.
 
In any case I was referring moreso to the RROD issue on the Xbox with my first paragraph, especially since that was the most prolific issue and was indeed a symptom of more than just a switch in the solder used.

You can think of it two different ways that mean the same thing.

1. Microsoft and most electronics companies didn't design their electronics correctly with regards to lead-free solder. This include Nvidia, Sony, Toshiba, Samsung, LG, Sharp, Philips, etc. Once everyone was forced to use lead free solder in a relatively short period of time, there was a massive increase in the failure rate of electronics across the board. Some companies did relatively better with it than others. AMD, for example, did a better job than Nvidia, but even they had some early teething problems.

2. Basically none of the above would have happened if lead free solder hadn't been forced onto companies. Companies knew well the tolerances and behavior characteristics of leaded solder. Not so much with lead free solder. Even now most of the budget electronics companies continue to have problems with lead free solder when they didn't have those same problems with leaded solder.

Anyone that has worked in the salvage/recycling industry will be able to tell you about the massive influx of electronics when the industry was forced to lead-free solder.

So, could MS have designed the X360 better to accommodate lead free solder? Possibly, but as mentioned no one had a really firm grasp on all of the potential behavioral characteristics of lead free solder. And even now, cheap electronics manufacturers still can't reliably make electronics using it.

IMO, the forced switch to lead free solder was one of the costliest and most environmentally bad government mandated things ever to have been forced onto corporations and consumers. A better disposal and recycling policy would have been safer, less costly to consumers, and easier to implement. But you don't get a lot of political points doing that. It's much easier to get political points by pointing and saying Lead = Bad, even if the environmental impact from leaded solder in electronics was absolutely miniscule.

As to the PS3 versus X360? They all appear to be failing at the same rates. Just X360's that were going to fail would fail sooner due to the higher thermal stress (my launch 360 still works fine). PS3's on the other hand are failing later in life. If you use this forum as a microcosm of this. While you no longer see people with launch 360's failing (like mine), we've seen a large increase in launch PS3's failing due to the solder (hence reflow trick to get them working for a bit). Had both companies been able to use leaded solder, failure rates would likely have been right along the industry average pre-lead free solder.

Regards,
SB
 
The 360 didn't fail because they had to use lead-free solder, it failed because they didn't delay the 360 to rework it for lead-free solder, and it was a new process at the time that nobody invested in until they were hopefully forced too. The reason they didn't invest in it is because toxicity has no monetary impact. Everyone was in the same boat. Just look around you, there's no other consumer electronic that failed so spectacularly. Some products were delayed to be compliant. We have 400W GPUs using lead-free solder and they don't fail. It's mature now.
IMO, the forced switch to lead free solder was one of the costliest and most environmentally bad government mandated things ever to have been forced onto corporations and consumers. A better disposal and recycling policy would have been safer, less costly to consumers, and easier to implement. But you don't get a lot of political points doing that. It's much easier to get political points by pointing and saying Lead = Bad, even if the environmental impact from leaded solder in electronics was absolutely miniscule.
That's complete BS. RoHS was written specifically to make recycling easier and safer. Japan doesn't force RoHS, but they all switched to lead-free anyway because they have strong recycling policies. When you have to pay for the recycling of your toxic crap, all of a sudden you stop producing them with toxic materials.
 
That's complete BS. RoHS was written specifically to make recycling easier and safer. Japan doesn't force RoHS, but they all switched to lead-free anyway because they have strong recycling policies. When you have to pay for the recycling of your toxic crap, all of a sudden you stop producing them with toxic materials.

Japan switched to lead free otherwise they couldn't sell their products abroad. Some of the products that were sold only within Japan continued to use leaded solder for a while. With traditionally low failure rates. Even lower than currently possible with lead free solder.

One nice thing with lead free solder is that it has lead to an incredibly large boom in refurbished electronics.../sarcasm.

And yes, thanks for indirectly confirming that if they had not been forced to use lead free solder, then the X360's wouldn't have failed in anything other than normal numbers. And likewise we wouldn't have a steady stream of launch PS3's failing for the past few years now due to the same issue.

Regards,
SB
 
I thought recycling meant recycled materials, not refurbishing?

It basically is lumped in with what many consider recycling in the recycling/salvage industry.

Bottles and jars, for example, can also be "recycled" by reusing them rather than melting the glass down and making new bottles and jars.

So there's quite a few things which end up at recycling centers or salvage yards that just get sold and reused. Or refurbished. Or to be broken down for base materials.

From 2002 to 2011, I worked 3 months out of each year at a family metal scrap yard in Japan refurbishing computers and some electronics. The influx of those items rose quite dramatically when companies started having to use lead free solder. I did it just because I hated to see otherwise useable computer components go to waste. Although during this time, the Chinese bought as many working computer parts as they could from the salvage yards. So things would have gotten reused either way. But I liked putting together systems and then selling them cheap to Japanese families that couldn't otherwise afford a computer.

Regards,
SB
 
I know precisely what you mean here. :LOL: It is obnoxious, I agree.
Well -this is also a reply to what @AverageJoe said about what is said here is considered gospel-, the blitter rumour actually turned out to be true, call it intuition, call it foresight, forward-looking approach, view of the future, etc...

This forum has a reputation that precedes it for a reason. I am not saying that everything that has been said here is necessarily true, but I am going to trust people here more than in other places for obvious reasons.

Sure there's going to be associated costs and any decision is going to be risk vs reward just like adding ram. I wouldn't be surprised if they were an unknown at MS at this point and they are just targeting AMD's lowest bin.
I don't know if the final specs are going to change tbh, most likely not much (if at all), but contrary to Sony this time around, what MS didn't expend on hardware they expend on the R & D department for a highly customized hardware.

Hence the system, ironically, does seem more difficult to modify without deep consequences.
 
It basically is lumped in with what many consider recycling in the recycling/salvage industry.

Bottles and jars, for example, can also be "recycled" by reusing them rather than melting the glass down and making new bottles and jars.

So there's quite a few things which end up at recycling centers or salvage yards that just get sold and reused. Or refurbished. Or to be broken down for base materials.

From 2002 to 2011, I worked 3 months out of each year at a family metal scrap yard in Japan refurbishing computers and some electronics. The influx of those items rose quite dramatically when companies started having to use lead free solder. I did it just because I hated to see otherwise useable computer components go to waste. Although during this time, the Chinese bought as many working computer parts as they could from the salvage yards. So things would have gotten reused either way. But I liked putting together systems and then selling them cheap to Japanese families that couldn't otherwise afford a computer.

Regards,
SB

That is very nice of you!

Though I think we should also appreciate the aspect of consumer safety with regards to lead and waste products.

I have no numbers and this is just conjecture, but I think its great that leaded solder was banned for consumer use. When these products are thrown out, even when there are regulations surrounding this as I understand, some of these products may make their way into trash or sewage which may contaminate water.

Given the capacity for even small amounts of lead to cause anemia and neurologic sequelae, I think we should in the end be happy that lead products are banned :)

I think health and safety is ultimately what is most important when it comes to consumer products and public health :)
 
That is very nice of you!

Though I think we should also appreciate the aspect of consumer safety with regards to lead and waste products.

I have no numbers and this is just conjecture, but I think its great that leaded solder was banned for consumer use. When these products are thrown out, even when there are regulations surrounding this as I understand, some of these products may make their way into trash or sewage which may contaminate water.

Given the capacity for even small amounts of lead to cause anemia and neurologic sequelae, I think we should in the end be happy that lead products are banned :)

I think health and safety is ultimately what is most important when it comes to consumer products and public health :)

I couldn't agree with you more :)
 
That is very nice of you!

Though I think we should also appreciate the aspect of consumer safety with regards to lead and waste products.

I have no numbers and this is just conjecture, but I think its great that leaded solder was banned for consumer use. When these products are thrown out, even when there are regulations surrounding this as I understand, some of these products may make their way into trash or sewage which may contaminate water.

Given the capacity for even small amounts of lead to cause anemia and neurologic sequelae, I think we should in the end be happy that lead products are banned :)

I think health and safety is ultimately what is most important when it comes to consumer products and public health :)

I understand it's bad, but.

If leaded solder is so bad why is it still used in hobby electronics :/ ive got a tonne here still, and it was even recommended to me for use. Shouldn't they have switched by now?.
 
and you do not know me

if he's not the Papa or amd/nvidia chief, he can be wrong as all the others. included me and you.
For someone who's 201 post have all been as part of the MDF, I would have thought you'd read enough other posts to know he worked at Microsoft and on the "Durango" specifically.

He knows more than anyone about it that posts here, maybe you should listen to what he's telling you instead of going for the "he's saying it's not that good, I must go on the attack" which seems your default approach.
 
Status
Not open for further replies.
Back
Top