Today Intel moved ahead of both the NASDAQ and ARM on a 30-day comparison. And the gap has gotten wider over the last 4 days.
ARM's 77 cent drop today can only be viewed bearishly as it continues to support the conclusion that ARM investors are exhausted as the stock has tried and failed three times over the last month to break in an upwards direction.
And there are plenty of macro dynamics supporting an ARM move down. ARM had few major smartphone announcements at MWC 2013. It needs continuous product refreshment in order to maintain its huge P/E ratio. Apple is struggling. Nvidia won't have Tegra 4 products until the end of the year and has only one design win.
ARM's fabrication roadmap ends after 20nm and 20nm is thought of as damaged goods because it is planar and not FinFET. Intel is preparing to open up a two process node lead over ARM.
Couple this with significant insider selling and you have to ask what has been holding the ARM stock price up for this long?
Intel has broke above a resistance level at 20.5 or so. The chart looks like a clear double bottom. Based on the chart, it looks like it is going to test 23 based on my chart work. ARMH is struggling but I would not call the stock broken. It looks to be forming a double top with a neckline at the 42 level. If it broke that, I would feel more comfortable shorting it.
I also have a gut feeling 44.47 will prove to be the very top for the foreseeable future.
Any ARM fabrication delays are not really a problem from a cost/size POV as their chips are naturally small anyway. The issue is transistor performance and power for their designs, their A15 will need sub 20nm FinFETs to behave like a proper successor to A9 because at the moment it looks like a netbook chip that they are trying to shoehorn into tablets and phones.
'I also have a gut feeling 44.47 will prove to be the very top for the foreseeable future.'
Confirmed really. That ARMH peak looked very unstable at the time which is what mountain tops look like ;-).
The stock went really bearish today, not only did it drop below its 50 DMA but it also lost the strong support it had at 42 meaning it now has 2 resistance points above it as well as the 50 MA. Not good for the ARM hypers really especially as their Calxeda multi-node server looked pretty ordinary today, time to invent some new fantasies (and hook some new suckers) as the current ones have been found out ;-).
BULLISH Probability - 1 Day Price change extreme down, may pause/reverse
BEARISH Probability - Intermediate trend possibly bearish, Uptrend turned sideways, may continue or
Confirmation - Strong bearish 3 day chart pattern with Strong 3 day distribution.
Confirmation - Extreme bearish 1 day moneyflow
CURRENT PRICE 41.31, just above support, 40.61 ± 0.61, type Single, strength 8
RESISTANCE ABOVE +2.4% at 42.3 ± 0.63, type Triple, strength 10
+6.1% at 43.83 ± 0.66, type Double, strength 6
-1.7% at 40.61 ± 0.61, type Single, strength 8
-7% at 38.42 ± 0.58, type Single, strength 3
-10.3% at 37.07 ± 0.56, type Double, strength 6
-12.3% at 36.24 ± 0.54, type Single, strength 3
-14.3% at 35.41 ± 0.53, type Single, strength 1
-18.4% at 33.7 ± 0.51, type Double, strength 6
-22.3% at 32.11 ± 0.48, type Single, strength 8
-30.8% at 28.57 ± 0.43, type Double, strength 6
-32.7% at 27.8 ± 0.42, type Triple, strength 9
-35.4% at 26.68 ± 0.4, type Double, strength 10
-39.4% at 25.04 ± 0.38, type Single, strength 8
Wally said ARM's Fab problems were going to sink it in 2012........when the stock was priced at 22. He and his mystk7 alias work 80 hours a week at the used car lot trying to scam old ladies into buying Lemons.
The issue is transistor performance and power for their designs, their A15 will need sub 20nm FinFETs to behave like a proper successor to A9 because at the moment it looks like a netbook chip that they are trying to shoehorn into tablets and phones
The A15 was designed for the 28nm node, while the new 64bit cores where design for 20nm. You will get some cross over.
A single A15@1.8Ghz going full pelt on 28nm will draw just over 1.5 watts. Easily getting it with the budget of a phone. Clearly, running multiple threads/cores will blow the power budget, but this is just like existing Atoms of today (with the turbo boost).