Plinko
Wildcard berths that can't beat teams without a winning record should have homefield advantage
Yeah, I'm a bit puzzled as to why this wouldn't be a huge announcement.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Xbox One hardware breakdown by ExtremeTech.com after HotChips reveal
- Thread starter ekim
- Start date
Yeah, I'm a bit puzzled as to why this wouldn't be a huge announcement.
I said take it as you wish.
I dunno they droped the upclock on gpu randomly so perhaps they dont care. Mabye they are saving it for another event
And one mystery is still unsolved:
What's up with that black arrow between the CPU and eSRAM?
If you look to the 'special processors' on the left, a number of the processors listed relate to the Data Move Engines.
I guess the black line signifies that data can be copied by the DMEs through the attaching bus to the CPU. It's not represented as a bus in and of itself because that would confuse people into thinking the CPU has direct access. But they want to have 'something' to show there is a way to copy data from eSRAM to the CPU.
And that's how I took it.
It seems I wasn't the only one either.
My point still stands though. These boxes that do all these magic multimedia functions aren't worth a damn in the marketplace. No one is going to shell out $100 bucks let alone $500 bucks because it can control some other things on your multimedia rack.
People have been using universal remotes for a long time and even plain ass remotes for even longer--people like it that way. The sales of these multimedia boxes prove it.
This seems so very off. Do you get up to do these things now on your 360/PS3? No? I didn't think so. A $500 dollar Xbone packed with multimedia features isn't going to change that. You're still going to swap from game, to Netflix, and now Blu-ray (did you forget it's in the Xbone now? You can bet anyone with blu-rays will make PS4/Xbone their new player of choice) without having to get up...same as you did before.
I mean, I love fancy gadgets and new ways to interact with my media just as much as the next guy...but let's not blind ourselves like the Xbone is doing something fancy and revolutionary.
boredandlazy
Member
This would be much more significant than that. This is something you'd tell your potential customers ASAP. It would be something you'd proudly proclaim as an advantage over your competitor. Unless they somehow were hoping Sony didn't find out?
SEGAvangelist
Member
PAX? TGS? Who knows? MS is hard to figure out these days.
phosphor112
Banned
Lmfao. The shit that guy writes.
Holy fuck. Lmfao.
SEGAvangelist
Member
Dude is amazing. Very entertaining.
walltowall
Member
I see your point but there are quite a few people who carried Sony when the PS3 was $599 buying it solely for the fact that it was the best bluray player, hell the thing even played SACD. So it is fair to say that additional features do help sell a console. I don't think that attach rates on the PS3 were lower because everyone reads Digital Foundry, but rather gaming is a secondary feature on many - not most - PS3s.
The prospect of how Kinect V2 will affect the buying decisions of families who already own V1 is interesting since Kinect is the first console add-on that hasn't been shat out and forgotten.
I guess we'll all find out by next Xmas.
Personally I'd take an AMD project manager (including the 7770 and 7790) view point over any post in this thread. But thats just me, call me crazy.
It seems a bit speculative and I'm not sure the sequence of logic used to conclude that - that there'd be no upside to an arbitrarily clockable interface - is bulletproof. Would welcome more expert opinion, but such an interface to allow another - lower - clock would be useful if it was intended to reach a power consumption/thermal target? From a pure performance point of view there might be no reason for it, but that's not the only consideration.
That's not to say it isn't the right figure, just not sure there's no reason they'd want to temper that speed, as the writer assumes.
As for why they wouldn't announce it if it was the case - it's the same reason they wouldn't announce 1.5Ghz, or why Sony hasn't announced PS4's CPU clockspeed. Because in either case they're unlikely to exceed 2Ghz, and even something like 2Ghz - to the untrained eye - would raise questionmarks in simple comparisons with their last machines (3.2Ghz).
They would if it played games too. Why do you keep saying "no one"? I might even pick up two Xbones.
I very much disagree.
What propped the PS3 up in those early years wasn't the secondary feature of the blu-ray player (which was a bonus and very much not-supported very well from the get go--like most new standards are) but it was the Playstation brand name itself.
Uh, did you skip coldfoot's post? Hell, he even questioned the buying public's want for a fancy multimedia box before I did.
As for Dave, you can believe him if you want..but that quote has been posted plenty of times before you by the infamous SenjutsuSage and a few other rejects from Beyond3D. If it had any significance it would have been heavily reported on. I'm not even sure if he's still with AMD and if he is, he's most certainly not working on either of the consoles.
I used the word no one because it was in context with previously released media boxes.
There's nothing "new" that Xbone brings to the table in multimedia functions and people know it--I mean, it can't even control the DVR.
Seems unlikely since their formula does not take the protocol overhead for ensuring coherency into account. In addition, since the PS4 has the same 30GB/s of cache-coherent bandwidth it would also be clocked at this speed. However, in both cases the leaked documents consistently point at 1,6Ghz although in both cases the 30GB/s figures was also already present.
I really can't follow the math used there - I have an idea but this doesn't work out for the result of 1.9GHz.
I'm pretty sure that was pre-Hotchips presentation and coincidentally only really reported on that crazy ass blog we were talking about a little while ago.
http://misterxmedia.livejournal.com/132494.html#comments
In the comments it's linked to the video.
Their formula is 1,9Ghz * 256bit / 2 (for a clock ratio of 1:2) = 243,2 Gbit/s = 30,4GB/s (Hence the ~1,9Ghz to come to 30GB/s)
But as I said, cache-coherent busses have protocol overhead.
Good points.
Anyway, as I edited, I don't see anyone publishing CPU clockspeeds whether they're 1.5 or 2Ghz. To people trained to compare chips by clockspeed, any of the possible candidates will look 'bad' next to the clockspeeds in their last gen systems. That's probably why they're happier to just say 'a 8-core x86'.
phosphor112
Banned
You specifically clicked onto a link to lead you to a thread about the technological aspects of the Xbox One. In short, a tech related thread. People started comparing technology to it's nearest competitor... so you come out and say "Well, Xbox One has the games I want." Paraphrasing of course.
Let me fill you in on something.
This isn't the thread for that. No one gives a shit unless they wanted to talk about it in the first place. It's shit like your post, DBZ shit and other garbage that derail threads.
Thanks.
McHuj
Member
It does? The vgleaks seems to say that the coherent Onion/Onion+ bus's share around 10GB/s and there's less then 20GB/s for CPU to memory (non coherent with the GPU).
statham
Member
1.9!! Thats another 5-6% gain!
edit: I remember on the 3dforums they were saying the xbox cpu was faster months ago.
Cache coherent is not a condition to be called a processor. I doubt the Xbone APU has some of the features present in my motherboard as well ..
MS logic of calling every trivial block as processor.
Charlie has also been vehemently saying that the Xbone APU yields are disastrous.
I believe what he's says implicitly and with very good reason.
Afaik the PS4 only has 20GB/s of cache-coherent bw?!
http://www.eurogamer.net/articles/digitalfoundry-how-the-crew-was-ported-to-playstation-4
I think we ALL knew you'd show up here for that tidbit.
Something that no one has reported on and is something significant that MS would have released immediately--which means it's probably inaccurate.
You know he's citing an article from May that's 100% guesswork, right?
statham
Member
I said, they have coherent access to the same MMU
You should look up the definition of a processor
Yeah.
I do, in fact, remember that.
I am just saying that Microsoft would not price their system at $100 more, which I understand includes the pricing of Kinect, and then put themselves at a huge disadvantage technically that would show itself so soon (as in third party titles this year or next).
The two systems will be similar enough that this is not the case even for a long while.
Will 1st party Sony titles look better within a few years, like everyone's nerd savior Mark Cerny said, sure. I just think third party titles should remain fairly similar. Also, I think MS 1st party titles will also be better when they squeeze as much as they can out of the Xbox One.
Yes, but it's hardly a justification.
And you should look around at how many vendors call every trivial block a processor ..
walltowall
Member
Sony thought exactly what you wrote. Gamers would get a second job so they could drop $600 on a PS3 just because of the Playstation brand. They were proven incorrect.
hugs ToyBroker
<3I think you're misunderstanding me. I was saying that brand loyalty as a result of PS2 (not secondary features like blu-ray or linux) is what kept the PS3 barely propped up in those early years...not the runaway success that Sony thought it was going to be at the time.
So if they decided to write "special purpose blocks" it would've been ok for you?
phosphor112
Banned
Nope. I can see that MS would up the CPU. It's not hard. I expect Sony to upclock it as well. Both of those are very low power processors and are much more likely to see clock increases over the GPU (since that would produce a lot more heat).
Those are the maximum numbers for individual pathways. The important figure is the total maximum cache-coherent bandwidth (with respect to the CPU caches). My take on the info was that the ~20GB/s shown in the Orbis figures correspond to the 20,8GB/s on the Jaguar modules in the Durango figures, and that the 10GB/s are additional cache-coherent bandwidth for the GPU. That reasoning is also based on the assumption that both systems have the same Jaguar cores which, hence, have the same interconnect to the MMU.
The thing is that the Durango figures show the maximum bandwidth of individual pathways without showing interdependencies. Another example is that part of the figure that shows the move engines, where every single move engine is labeled with 25,6GB/s. The total bandwidth, however, is not the sum of those numbers, but also 25,GB/s.
Whoa, somebody woke up on the wrong side. My contention isnt with the thread, but those who choose to side track it into a pissing match. I don't recall the OP saying "here are some Xbone specs now let compare and contrast with PS4 for the millionth time. Chill out though, It ain't that serious bro.
In the semiconductor industry, cost goes up exponentially with die size, so when you're talking about costs, the physical makeup of each component matters.
MaulerX
Member
Found it! This is where I read it: http://venturebeat.com/2013/08/26/m...se-details-are-critical-for-the-kind-of-expe/
phosphor112
Banned
Pissing match?
Whenever I see these technical threads where the XBone is just getting bashed with all this technomombojombo I think, "well what does this mean for actually games".
Then I'm reminded...
Tell me more. Please.
The PS4 is a perfect piece of hardware to compare considering how similar those are. The article in question mentions Onion and Garlic, two buses that are actually used in the PS4.
It's a very valid discussion.
What isn't valid, is your PR poster that you decided to embed into your post.
Once again, thanks. It's appreciated.
afaik neither Sony nor MS confirmed clock speeds for their CPU so far.
They've said nothing about CPU clock.
It might be, might not be. It doesn't seem to be as conclusive as the article suggests.
Holy crap phosphor112 laying the smacketh downeth.
My god...that's their chip architect? Shudders :X
My god...that's their chip architect? Shudders :X
GameSeeker
Member
At the moment it's just a guess by Semi-Accurate. They don't have any verification.
dragonelite
Member
Any words what GPGPU API the X1 will use?
From what i gathered the DirectCompute is lagging behind massively compared to Cuda and OpenCL.
Or will they use C++ AMP? Im not sure if its build on DirectCompute or can we expect a massive api update for DirectCompute this or next year?
And is there anything said about the API list for the X1(Win32,WinRT blablabla).
From what i gathered the DirectCompute is lagging behind massively compared to Cuda and OpenCL.
Or will they use C++ AMP? Im not sure if its build on DirectCompute or can we expect a massive api update for DirectCompute this or next year?
And is there anything said about the API list for the X1(Win32,WinRT blablabla).
Similar threads
- Locked
- Poll