Sample is up over 12,000 units now: http://demo.hiraku.tw/CPUIdentifier/
Ironically, if anything, those results make it seem that Samsung has more business from Apple than that 2013 article indicated. They stated Samsung would get 30-40% of the 2015 new iPhone business, but Samsung has 41%.
P.S. There is no way I'm going to run that app on my wife's iPhone. I'm glad 12000+ other people are happy to be guinea pigs.
The source code is safe, I checked it yesterday, but yeah, when I get my 6s, if I ever feel like knowing which manufacturer I got, I´ll just use the 5-6 lines of code that actually fetch the CPU code and run it myself.
Do you have a link to the source? Or did you just disassemble the app?
Edit: apparently the original author released the source and not just the alternate version. Thanks Arachnotronic.
They left some wiggle room at the end of the article:Ironically, if anything, those results make it seem that Samsung has more business from Apple than that 2013 article indicated. They stated Samsung would get 30-40% of the 2015 new iPhone business, but Samsung has 41%.
What better incentive to deliver the chips fast than getting a bigger piece of the pie? :awe:There is also a possibility that TSMC and Samsung would equally share Apple's chip orders for 2014, the sources noted.
Sorry but too many people are making elementary mistakes. 2500 samples from 13 million units sold in 3 days. Thats a 0.0192% sampling rate. There is no way you can arrive at a correct conclusion on TSMC and Samsung ratio with such a low sampling rate.
2500 is more than enough to get a useful estimate, in fact at a 95% confidence level your confidence interval will be just +/- 1.9%. At a 99% confidence level it would be +/- 2.5%.
2500 is more than enough to get a useful estimate, in fact at a 95% confidence level your confidence interval will be just +/- 1.9%. At a 99% confidence level it would be +/- 2.5%.
If we assume that for the purpose of this discussion a +/-5% confidence interval at 95% confidence level would be sufficient, then we would need less than 400 samples.
True, but to be fair, your statistics figures are assuming a perfectly random sample. For all we know, the results could be very different by region. What are the percentages by region? How many of the samples are from each region? Etc.
These details do matter, and as far as I know, we don't know them yet.
It looks like A8 IPC already is close to Haswell.I get a ratio of 5548 on my 4770K @3.9 GHz, compiled with gcc 4.8.3.
...
When forcing the frequency of my 4770K to 1.9 GHz 176.gcc gets down to 2769. When
setting it to 1.3 GHz I get 1887. Apple A8 @1.4 GHz is getting 1810.
All statistics has to assume a random sample, otherwise any and all data is useless.
As far as the sample being biased, it is a possibility, but unless there is actually any evidence of it, the normal thing would be to assume that it is not.
Someone at Realworldtech posted some SPECCPU 2000 176.gcc result on a 4770K with frequency scaled down.
http://www.realworldtech.com/forum/?threadid=154177&curpostid=154246
It looks like A8 IPC already is close to Haswell.
True, but to be fair, your statistics figures are assuming a perfectly random sample. For all we know, the results could be very different by region. What are the percentages by region? How many of the samples are from each region? Etc.
These details do matter, and as far as I know, we don't know them yet.
The CLT is responsible for this remarkable result:
The distribution of an average tends to be Normal, even when the distribution from which the average is computed is decidedly non-Normal.
Thus, the Central Limit theorem is the foundation for many statistical procedures, including Quality Control Charts, because the distribution of the phenomenon under study does not have to be Normal because its average will be. (see statistical fine print )
Statistical fine-print: The distribution of an average will tend to be Normal as the sample size increases, regardless of the distribution from which the average is taken except when the moments of the parent distribution do not exist. All practical distributions in statistical engineering have defined moments, and thus the CLT applies.
Where does this 70% come from? And just a number of days ago you said TSMC got the iPhone 6s.The reality is the distribution itself is not static over time, as Samsung improves their yields then so too will their share of the A9 pie increase. They were contractually awarded 70%, the ratio is only lower (for now) because of the yield disparity at this time.
Where does this 70% come from? And just a number of days ago you said TSMC got the iPhone 6s.
You also said that the Samsung fabs that would produce the SGS6 were empty, yet 14nm is in the S6.
Your track record for this kind of rumors isn't very good, TBH.
My post wasn't an ad hominem. I don't have an agenda.I really don't know what you are going on about. If you can't follow the conversation, then why make it into a thinly veiled personal attack?
I've explained numerous times the differences between expectations and realities in the fabs and the market right now, if you (or other characters) want to pick and choose from my posts the bits that only support your preferred agenda then please, by all means, go for it.
Doesn't change reality one iota, just makes it more clear to me that some folks have veered into becoming obsessed with something that has little to do with technology.
It has gone up to 26.000 entries... and the ratio is slowly shifting towards a 50/50 split.Still same numbers. So pretty much 60% TSMC 40% Samsung confirmed.
It has gone up to 26.000 entries... and the ratio is slowly shifting towards a 50/50 split.
To top that off, if we use IDC's argument that "statistics don't necessarily reflect reality", the split might actually be closer to 40% TSMC 60% SS, and we can see that IDC's projections about the current state of non-intel foundries are at best educated guesses, and at worst purposeful misinformation.
I choose to interpret it as incompetence vs purposeful deception.
At any rate, for those who are curious, at the moment its all TSMC. Samsung was originally apportioned 80%, then 70% due to softening confidence in their yield ramp rate, and that was later on further reduced to 50%...but their yields are so low right now (compared to TSMC's) that practically all the A9's headed to market on the eve of the iphone 6S release were fabbed by TSMC. Should Samsung get their yields up, and same goes for GloFo for that matter, then they'll get to claw back their portion of the volume contracts for the A9 in the coming months and quarters.
It has gone up to 26.000 entries... and the ratio is slowly shifting towards a 50/50 split.
At this point it has become increasingly clear to me your main interest in this discussion is to call out another forum member on the basis of some speculation he made. It doesn't matter you yourself have been patently wrong on this very thread, you keep hammering at it to show us how bad the other guy is. Well, it's your right to prove someone wrong on the basis of facts, so don't let me stop you. Meanwhile though, I do believe the other forum members should be aware of how you choose to interact with the same forum member:To top that off, if we use IDC's argument that "statistics don't necessarily reflect reality", the split might actually be closer to 40% TSMC 60% SS, and we can see that IDC's projections about the current state of non-intel foundries are at best educated guesses, and at worst purposeful misinformation.
I choose to interpret it as incompetence vs purposeful deception.
Oh look, a page long post by idontcare implying he knows the answers to our questions but offering no information
I can see how you became an "Elite" member, you're clearly way too smart and too important to even bother to answer us "fanboys".
Why is there even an overclocking and CPU forum when we can all rely on the benificent and all knowing genius that is Idontcare?
I don't know if you have some personal history with IDC or you just think taking on a 'big guy' in the room will earn you rep fast, but you might want to consider making your posts more about technology and less about forum members. It will make a far more attractive and productive conversation, after all this is why most of us attend this forum.I have no idea why anybody listens to you, you clearly are out of the industry.
I don't know if you have some personal history with IDC or you just think taking on a 'big guy' in the room will earn you rep fast, but you might want to consider making your posts more about technology and less about forum members. It will make a far more attractive and productive conversation, after all this is why most of us attend this forum.