The Xbox One is jaguar cores, unfortunately.
The most important measurement must be idle. There is near a 3x reduction on consumption. Wow!
Porting a CPU layout from a foundry process to another one
is not , or rather wasnt , economicaly viable because you ll
had to redesign the complete circuit in function of the new
process transistors caracteristics but if they managed to do
so with Jaguar means that their computerized tools are very
advanced and work well so they ll need far less workforce
for implementing their future designs.
Again, something is gravely wrong with that chart. A i7 quad + SCREEN idles at a similar power usage (often lower).
Hell, many notebooks browse the internet + Screen at that level of power consumption (typical craptop with 48 watt hour battery can get about 3 hours browsing 48/3=16 watts for light load and screen).
And looking back on AMDs product mix, their mobile bobcat shipments are dropping fast. But again, so is everything they make.
then I would assume Something is wrong with your testing methodology then...notebook check doesn't have many notebooks with idle less that 4w...
Notebookcheck measures with the screen.
Its quite good power wise but 2w is a fairly small difference (but still quite noticeable). Note the difference as a % of the whole will become smaller in real life because this test is without the screen.
http://www.notebookcheck.net/Review-Update-Dell-Latitude-6430u-HD-Ultrabook.92936.0.html
Newest review on notebookcheck with an i5 ULV.
Power consumption idle.
(note that notebook idle powers are idle/office work,etc./internet).
7.8/10.9/12.2
I thought we were talking about your i7 idle...but even though normal workloads are on average 2w less, at max load is up to 15w less for kabini - 26w vs. 35w. If they had the same power envelope then maybe the performance would be more competitive with i3 at the list.
Notebookcheck measures with the screen.
Its quite good power wise but 2w is a fairly small difference (but still quite noticeable). Note the difference as a % of the whole will become smaller in real life because this test is without the screen.
http://www.notebookcheck.net/Review-Update-Dell-Latitude-6430u-HD-Ultrabook.92936.0.html
Newest review on notebookcheck with an i5 ULV.
Power consumption idle.
(note that notebook idle powers are idle/office work,etc./internet).
7.8/10.9/12.2
It can be found in a PDF that was released at time
of bobcat launch with estimations of segments future evolution.
The paper clearly show the 200-500$ segment as being
the faster growing and targeted one.
http://www.amd.com/us/press-releases/Pages/amd-fin-analyst-day-2010nov09.aspx
No need to be fluent in english to understand thatLol, a 2 year old marketing slide of estimated projections is your proof!?!
I know english isn't your native language, so let me help you here http://dictionary.reference.com/browse/proof?s=t
You are saying AMD is limiting themselves to a max price target of $600.
Do you have proof of any of this?
Not really relevant the x202e notebook they tested has horrible throttling problems (like 1.2 ghz core and 300-500 mhz igp under load). Difference in performance should really only be around 20% at most.
Toms review on the same or similar cpu (i3 ULV possibly different model--100 mhz difference)
Its quite good power wise but 2w is a fairly small difference (but still quite noticeable). Note the difference as a % of the whole will become smaller in real life because this test is without the screen.
Best to compare Kabini at its best (1600 single channel) to i3 at its best (dual channel), where you get 30 fps for Kabini and 40 fps for the i3. Not too bad.
Kabini would gain 15%, and the i3 would gain nothing?
i3 at its best = 1600 dual, tom used 1333 dual, so the i3 would also see a small increase, I would think, still both could the faster memory, and that's it, normally a memory bottleneck will benefit more the slower chip for a comparison.
but also I suppose Kabini is more starved for bandwidth considering it's only 64bit.
at least Tom's Hardware used the same software, memory and Hard Drive... a much better review than the other,
The i5 tested in the review had dual 1600 and got to 44 fps. Would the GPU performance of the i3 and i5 be the same under ideal conditions?
Now that I think about it, Jaguar is pretty much the best CPU out there that Sony/MS could have used for a console. It'd be terrible for PC gaming since you need more ST CPU power to overcome the huge overhead; but on a console that's not a problem. It's a balance between low power and performance. The low TDP of the cores lets you use the budget on what you really want - GPU power.
Not to mention low cost as well. Basically they used the cheapest lowest power cpu they could get away with and devoted as many resources as possible to the gpu.
Not just lowest cost- lowest power usage, too. From the reviews of Kabini it looks like Jaguar has excellent efficiency, letting them use more of their (very strict) TDP budget for graphics performance.
Kabini would gain 15%, and the i3 would gain nothing?
No need to be fluent in english to understand that
you are of bad faith and willingly changing the subject
with your ridiculous link.
You think that dedicated Socs can be develloped in 6 months ,
that such a company follow three months raodmaps..?..
You original question , though, :
It was a combination of three things - low cost, performance per watt, and general performance.
ARM can beat Jaguar at cost and performance per watt, but not in general performance (at 2Ghz Jaguar is faster than any ARM CPU AFAIK).
Intel can beat Jaguar at performance per watt (possibly) and general performance, but not at a low cost.
Also a factor is that they would have had to use a discrete gpu with Intel, which ultimately could have given better performance, but at higher cost and power and the inconvenience of dealing with 2 separate companies that are both rather challenging to negotiate with.
Not necessarily. Intel has the building blocks for an iGPU, they would just need to make a larger chip with more duplicates of the same EUs and with a faster memory bus. Double the number of EUs from Haswell GT3e, add GDDR5 (with their experience from the Xeon Phi), and you've got a console part. Intel could have made the part if they wanted to, but I imagine that they wouldn't accept a price which the console makers would be happy with.
Resume: AMD had both the technology and the price. Intel has none.
what about the xeon phi, couldn't that make an awesome graphics + compute system for consoles?You honestly don't think Intel has the technology to make an APU? Where have you been for the last 3 years?