Clover Trail+ versus ARM, the results may surprise you

Subject: General Tech | June 17, 2013 - 02:37 PM |
Tagged: arm, clover trail, tegra 3

ARM might be in for more of a fight than we had thought if they want to keep their market share for the next generation of cellphones, assuming of course that they are sold in North America.  The Register posted about research recently done contrasting performance and power efficiency between several phone CPUs; the Lenovo K900 with a 2.0GHz Atom Z2580, a a Samsung Nexus 10 with a dual core 1.7GHz Cortex-A15, a Galaxy S4 phone running a "big.LITTLE" Exynos Octa with paired quad-core Cortex-A15 and Cortex A7 and even a Asus Nexus 7 with an Nvidia Tegra 3.  Those phones give a good representation of current generation technology and it seems that while the performance for the top phones was very similar, Intel's new ATOM did it with 2/3 the amperage, specifically an average of 0.85A  as opposed to the 1.38A of the second lowest competitor.  ATOM seems to have finally found a market segment it can do very well in as long as the price is right.

View Full Size

"The industry analysts at ABI Research pitted a Lenovo smartphone based on Intel's Atom-based Clover Trail+ platform against a quartet of ARM-based systems, and Chipzilla's system not only kept pace with the best of them, but did so using less power."

Here is some more Tech News from around the web:

Tech Talk

Source: The Register
June 17, 2013 | 02:48 PM - Posted by Goofus Maximus (not verified)

The part in the Register article that gives me pause is: "ABI Research provided no details on the content and construction of their benchmarks".

Just what is ABI Research, anyway? Is this a credible firm, or one of those "sponsored research" facilities that make their money by spinning the numbers?

June 17, 2013 | 04:11 PM - Posted by Anonymous (not verified)

Was scanning Google News Tech articles and was hoping PCPER had an article of it`s own on Haswell E enthusiast chip.
Sounds awesome !
C'mon Ryan...this deserves a podcast ; )

June 17, 2013 | 04:14 PM - Posted by Anonymous (not verified)

IMO , when Silvermont hits ARM will lose market share.
I would be more inclined to buy something with Silvermont inside.

June 17, 2013 | 04:16 PM - Posted by Anonymous (not verified)

Intel Inside !

June 17, 2013 | 04:51 PM - Posted by Anonymous (not verified)

Intel Inside, money outside of your wallet!

June 18, 2013 | 09:58 AM - Posted by Anonymous (not verified)

It's the cost of the chip, most people only use their phones for non intensave browsing, for phone users that do, the phone's GPU is more of a factor than the CPU! Intel is still using their process node lead to try and push their x86 based solutions, but ARM and the other chip fabs are closing Intel's process lead! An ARM based phone chip paired with AMD graphics will probably make its way into the market, to compete with Nvidia, as well as Intel. AMD's hUMA unified memory technology will become an ARM based product paired with AMD graphics the will result in power efficient and powerfull phone/tablet APUs to keep Intel on the run! Watts used, and not amps without volts, will reveal the real low power winner!

June 18, 2013 | 02:36 PM - Posted by nobody special (not verified)

iNTEL won't have a modem until 2014 (integrated) and silvermont won't be in anything until 2014 also. By then it will be facing 20nm A15's so again will be a moot issue. Also note the smacking octa gave it in gpu. You will see the same thing and WORSE when T4 hits a phone as it's better than octa in gpu by quite a bit and also dwarfs S600's A320.

Not much will change at 14nm for Intel either, as it will be facing the next core from Arm and the rest as mentioned by another here are catching on in fabs tech with the consortium all working together to beat Intel. You can't out spend IBM/Samsung/GF/TSMC together on fab tech. The combined profits of these make intel seem like a small company. Samsung alone will likely break 20B (based on projections of S4 sales, over double last time at 40mil, this time looks like 100mil+, just sold 10mil in a month!) this year in profits and this last Q made 7B while Intel made what like 2.3B (will samsung break 25B this year??)? They will out spend you ALONE, never mind the group together. Even a fool should be able to do this math. Samsung can spend in ONE QUARTER on fab tech, the same as INTEL spends in a YEAR. How do you win that war? If I was samsung I'd put 10B per year into fabs until I win.

If apple wants to remain in this race they should start building 2-3 of the best fabs money can buy NOW, so they don't depend on anyone for memory or soc soon (and invest a Bil/yr in games for the next 5 years, that ONLY run on IOS - what's that 200 $5mil games that are exclusive? AWESOME, but they're too dumb to do this). They now depend on TSMC (ROFL) and will have late products like NV/AMD does with vid cards very year...GF is even worse which AMD is still totally tied to (and killing their die shrinks). But the consortium should bring TSMC back into line by 16nm (and should be near the same as Intel's 14nm, since most consider Intel's not really 14, but closer to 16). I think GF ends up out of business in the next 5-7yrs once ATIC is tired of investing in losses after losses. They really need to execute to get out of trouble, and they already delayed abu dhabi plant (which shows ATIC losing faith, as they cut funding until GF gets some results).

NV will be the one with odds on their side now that they will be heavily moving to samsung as they fill their apple losses with huge GPU's that are 3-4x the size of apple socs, so you don't need many to eat up all the lost fab wafers, not to mention the socs also. Apple should have very little trouble once NV is all samsung and no TSMC. Unlike TSMC (who is always late for AMD/NV etc), samsung has NEVER been late on HUGE apple product launches. I see very good chips for NV after T5 (which I believe is still TSMC). From 2015 on NV should have lots of samsung chips which will help their war with Qcom who will be fighting over TSMC with Apple. Apple already tried to buy ALL TSMC production but failed (but this is the kind of company Qcom will be fighting).

I don't see Intel producing a competitor's SOC, maybe other crap but not direct competition they are trying to take out with fabs. So TSMC should be able to extort some extra cash from both Qcom and Apple...LOL. "Pay us more money, or um...The other guy will get a better launch this time"...ROFL. Apple needs TSMC more than TSMC needs Apple. There is nowhere to go with no samsung in the picture (GF? You must be smoking something, Charter? Owned by the joke GF...LOL). If I was TSMC the apple contract would be ULTRA high cost. I'd say a price hike to Qcom at say 10-15%, and a good 30% or more premium to apple...Apple does this crap all the time to wireless companies, parts suppliers etc. Screw them while you can due to their own BS with lawsuits costing them samung. Time for the hand that feeds them to BITE BACK.

June 19, 2013 | 04:30 AM - Posted by Anonymous (not verified)

You say that 14nm intel won't matter? How so? Because it will remove 1/3 the power and allow for higher clocks with lower volts. No to mention next gen atom has 3d transistor, trigate tech and the new batch of gpu. Intel's processor market share is 2x that of ARM's. And Intel owns one of the highest stakes in TSMC therefor if they wanted to, they could screw ARM over completely. Intel's in the best position they could be.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.