patrick.net

Like I said, there are no other players to take out x86. Lots of people have tried, they've all failed miserably, even Intel tried with the Itanium, spending 10B on it, but no go.

Phones and small devices are different, because they can be uprooted, so the best player wins. Look at the consoles, they've all been fairly successful, and aside from the xbox360 have been using weird chips and chipset! Phones are the same, as long as a phone has a single OS working, it's good to go.

E-man says:

>>You might be able to pick this up for just above $335k.

But you realize that the 2nd lien of $150k does not just go away, do you not? If you buy at this auction, what you buy is a mortage and not a house.

As a side note, my Visa purchase @70 went through today, although only 50 shares. I'm not sure how low it might go before we hit a bottom, but I'm guessing it won't drop by too much more, even if it follows the market. Next month, I'll see if I can pick up another 50-100 shares or look at CCME instead. Thanks for the heads up on that stock as well.

as long as a phone has a single OS working, it’s good to go.
And that is no small feat as you try to shove more functionality into the phone. Definitely an interesting space to watch as customized RISC chips (most ARM cores) currently own the space. Intel is trying to drop in from above with an x86 architecture (optimized for power consumption, which is where custom RISC chips eat their lunch). If they're successful, they bring all those apps that have bring written and compiled for x86 cores. A betting man, though, might say that it's Intel that is going to get taken out from below, by power efficient ARM cores (disruptive technology?). See this. There are a number of ways to create a power-efficient server chip for hyperscale applications like those that run at Google, eBay, Facebook, and so on. One is to rip damned near all of the guts out of a Xeon processor and make an Atom chip, which Intel has done. Another is to beef up a MIPS or ARM RISC processor aimed at fairly modest workloads so it can handle server workloads. A number of niche server chip makers have deployed MIPS designs, and now it is ARM's turn to take a crack at Intel.

EBGuy, no doubt advances in server CPU technology will percolate up from the low end and low power applications.

@EBguy
The difference between applications for the phone and applications for the PC are generally complexity. Porting applications and maintaining them on another architecture is massively time consuming. However, because the phones are essentially on ONE type of hardware it's relatively easy. The phone doesn't have a USB, printer, RS232, Soundcard, DVI, VGA, joystick port, etc. It has a couple of ports but not much else. There are hundreds of sound cards. Thousands of video cards. Thousands of mother boards. Thousands of 3rd party devices that all need to work in an OS. It's a massive project just for MS to maintain windows. The sheer number of combinations is incomprehensible. They have MASSIVE testing server rooms setup with thousands of computers to test windows.

Computers and phones aren't even remotely close. A phone is a simple device compared with a full blown computer.

AMD was involved in the phone market and has since sold that business. Intel has been in that business many times. A phone is interesting, but it's also a business that you can quickly and easily port everything to. 2 years, and all legacy stuff can be thrown out practically! Not so with computers.

x86 is probably one of the *WORST* computer designs out there. The original 8086 was horrible. The 286 made it worse. 386 even worse. 486/586/pentium4/amd 64 just added to the mess. But it became more and more powerful with each step, but everything is a nail to that processor. Itanium has some great stuff in it, but it never took off.

Same with server machines. Google might use X chip, but not many other companies are in a position to do that. It's expensive to make a chip. Even if it's faster and more power efficient than an x86, the x86 is here today, it's fast today, and it's incredibly cheap today.

CPU and Server chips will likely gain from these markets, but those markets aren't going to ever touch these cpu's. I've been waiting for 20 years. I've used all the 3rd party chips and loved them. The 68000 series was so incredible and I used it for a long time. But the x86 had the R&D due to the market share. It's so complex now, that no one has a chance to enter this market. Nvidia will fail spectacularly.

This market is mature though, and we have little need for more speed. More efficiency would be nice. Servers will drive this. Cores are important, so that will be pushed further as well.

pkennedy, I think we are (mostly) in agreement. I'll try to point out when (and if) we aren't, so we don't end up talking past each other (like engineers are famous for doing!)

A phone is a simple device compared with a full blown computer.
Mostly agree (especially the part about legacy driver support). At the same time, the OMAP processor (variants used in Droid, Nokia N900 and PalmPre) contains multiple, specialized processing cores (a general purpose ARM core + a DSP + audio). This degree of specialization (and integration) meets the needs of mobile computing market (read: power efficient), but does present some challenges when migrating to the next gen platform (have fun porting that DSP code!) I'm not sure how Android handles this (or if it even uses the DSP for video decode, etc.) but I know that Maemo (Nokia linux) has a habit of abandoning older platforms (see N770 -> N810 -> N900). But as you say, that's a virtue, not a vice as it gets us the latest goodies on a new phone/tablet! I think that open source is helping to take the pain out of specialized cores (allows companies to sell chips, not software).

Google might use X chip, but not many other companies are in a position to do that.
I know of a company that took great care in its hardware abstraction layer. They could (and did), fairly easily switch horses (CPUs) midstream depending on a best of breed strategy or target market. The code base could be easily compiled and run across multiple processor types (Alpha, x86, Mips). At the same time, driver support was limited to what was sold with the system and a limited number of options (much easier on the server side).

Speaking of legacy driver support, I recently Ubuntu-ized a relative's computer. I remember having a bear of a time trying to find the correct (internal, removable) wireless driver on XP for this system. Ubuntu didn't load the driver by default as it didn't meet their open source licensing restrictions (I think), but it did allow for a painless install under Hardware Drivers (went out to the intertubez and downloaded it). Very nice (never touched the command line)! No need to consult forums.wtf.windoze.hardware.com as I had to do for XP.

This market is mature though, and we have little need for more speed
Definitely agree with this; it's what Clayton Christansen calls technological overshoot. Time for that crappy ARM CPU to disrupt (or not?!)...

Nvidia will fail spectacularly.
You keep saying this and I have no idea what you're talking about. I've got to look it up (whatever IT is :-)

A single company can do what you're talking about. But they have to be massive in general and usually building their own software. You could put your web server software on any hardware now a days, but the cheapest bang for the buck is x86. Even if it's specialized. Specialized has the draw back that you never know if the company will remain solvent for the next generation. Anything you've "counted on" could be lost.

For the most part, anyone buying any software will stick with x86. Anyone developing their own will stick with x86. The small number in there will move to another platform because it works for them.

None of those technologies will disrupt the PC. While we don't fully appreciate the hardware, we also waste is horrendously with terrible code. There isn't a chance an ARM or other processor will replace the x86.

In a couple of years, many of the x86 patents expire I believe. Nvidia is working on their own CPU/GPU combo. I don't believe they'll get anywhere with it, and it will fail spectacularly. AMD is about to release their next generation architecture for servers, and mobile. Mobile will no only have x86_64 but will have a GPU built in, giving it the power to do fairly intensive graphics without the need of a video card add-in. This will reduce the number of chips in a system (lower cost) and reduce power draw.

Btw, I spend most of my time looking at new hardware and what it can do. I spend a lot of time trying to figure out how to get power usage down as well.

So all new hardware is important to me. Nothing so far has come close to an x86 based system in the last 15 years.

For the most part, anyone buying any software will stick with x86. Anyone developing their own will stick with x86.
This, as much as anything else, is what x86 dominance is about. A credible stack is needed to challenge the Windows (& Apple OS X) x86 program. I imagine a shift to "the cloud" could also change people's perception of what they need for usability. Here's an interesting tidbit, Moorestown (AtoMs targetting the mobile market), will be the first Intel chip to dispense with PCI bus support (polling kills battery life, you don't need it that space). On the handset (& tablet) side of things, MeeGo will be the one to watch for open computing (as opposed to the more restrictive iOS & Android). I agree that on the desktop (and desktop mobile space), x86 will continue to dominate.

Nothing so far has come close to an x86 based system in the last 15 years.
Cell is impressive (especially if you need the float point ops). Shocked to see the dominance of Cell and Nvidia's CUDA based platforms in the HPC (high performance computing) arena. Nvidia currently leads with FLOPS/watt (and add in per $ and per unit volume -- kills the AC bill, though).

It really comes down critical mass, not quality. Cell is impressive. Atom is impressive. Transmeta was very good! Via has good stuff. Itanium is great too. Risc has been out forever, and great. No one has even come close to a critical mass. Not even Intel could hit a critical mass with Itanium, which they spent something like 10B making.

Cell phones have switched processors/OS's/software every couple of years.
Consoles have swapped out processors every model. From the original coleco's, ataris, ninentdo's wii, xbox 360's, etc. The market can hit critical mass without any issues.

Cloud is "neat" but of little practical use for 99% of the population. I use it, it's neat. It's no disruptive technology right now.

The biggest threat is more software shifting to the browser, meaning as long as a computer runs a browser, any machine can be used. But we've had that with Java forever, and it never disrupted anything either.

x86 can't even get their new hardware instructions accepted and used! We're still building for pentium/486 based computers! Barely touching any of their "cool" instructions. x86 can't even get people to change their coding methods to utilize any of the new super powers they offer!

x86 isn't going anywhere. Not even AMD/Intel can influence the market.

CCME.

I'm just wondering what is going on with this stock, it looks to be on a long sliding path now. We're down to about 9.40 this week, with little change in direction for this stock.

I guess it could be said to just be following the market currently?

small stock, low volume, technical breakdown. Support at 8.60's and then 7's.

How often do you use technicals when pricing a stock and/or selling one?

I'm just wondering how technicals play into your buying/selling strategies, and how they play into value buying?

You'll have to know what you are dealing with. 33M shares outstanding and 350K avg daily volume (In this case, the effect of 3M+ dollars exchanged for a 300M+ market cap company is rather large) causing lots of price movement, tells you this is a techincal stock. 95% of the action enter and exit within a month. From an option writer perspective, it is really easy to manipulate. There are lots of other signs as well. In this case, I weigh technicals more than fundamentals. although I never buy a stock without solid fundamentals.

Buyer at 8.65 and on my short term watch list.

So a 1% trade per day is considered high? I guess I figured 1% ish a fairly relative number for many stocks. Maybe not for a small cap. I would have also assumed the higher number trading hands makes it harder to manipulate, or is it the fact that an option trader only needs a few million to move the entire stock, vs the high 1% turn over rate?

How long would you look to hold a stock like this, or in general any stock? Do you revisit the pricing very often and/or sell based on a quick jump in price? Such as rig, it jumped up 20% or so and I thought of selling, but then I figured I would be playing a buy/sell game of trying to capture short term profits? Or is this a viable strategy for me to look at, and just assume if I "miss" buy high by mistake one time, it doesn't matter because there is so much upside potential, that it's eventually going to go up anyways and I might just have to wait it out a bit longer?

When taking advice, I like to have a good idea of how a person works across the market. Taking one piece of advice without understanding the underlying strategy can lead to a disaster, so hence my many questions.

The biggest threat is more software shifting to the browser, meaning as long as a computer runs a browser, any machine can be used.
Agreed. That is what I meant when referring to the cloud.
It’s no disruptive technology right now.
That's what's so hard about disruption. By definition (Clayton Christensen), it sucks, but works in a niche market. And then it eats your lunch (and market share). Certainly will be an interesting market to watch and see if it gains a foothold.
x86 can’t even get their new hardware instructions accepted and used!
Insightful comment (cough, cough, as I run Ubuntu on my 10 year old laptop). That's what so interesting about the mobile market, as difficult tasks (video processing) do use the special purpose processors included in the SoCs for efficiency reasons. And as you pointed out earlier, the cellphone/mobile folks don't have to worry about 'legacy'.

"So a 1% trade per day is considered high? I guess I figured 1% ish a fairly relative number for many stocks. Maybe not for a small cap. I would have also assumed the higher number trading hands makes it harder to manipulate, or is it the fact that an option trader only needs a few million to move the entire stock, vs the high 1% turn over rate? "

It's relative, the point is it takes relatively little money to move the stock, hence it is more riped for trading.

"How long would you look to hold a stock like this, or in general any stock? Do you revisit the pricing very often and/or sell based on a quick jump in price? Such as rig, it jumped up 20% or so and I thought of selling, but then I figured I would be playing a buy/sell game of trying to capture short term profits? Or is this a viable strategy for me to look at, and just assume if I “miss” buy high by mistake one time, it doesn’t matter because there is so much upside potential, that it’s eventually going to go up anyways and I might just have to wait it out a bit longer?"

That's really tricky and not easy to answer and really the difference between making good money and not. If I knew the answer, I don't need to work as well. It's like playing poker, how to squeeze every last value bet and succeeding in this critical area. I just trust my experience. Since it is a techical buy, it should be a technical sell as well in the short term. Sometimes if I have a strong conviction about long term value, I keep a core amount as well. A lot of people plays the techincal and keep a core position.

As you can see in the stock market for the past tens years, buy and hold is not the correct strategy anymore. Making money requires quick buy and quick exits.

I'm just wondering how you would play something like visa/rig? They're value buys, but at the same time RIG took off and jumped 20% very quickly on me, which I would think would be a good technical time to sell it even though it's still in the value arena. I'm guessing you're considering Rig and Visa to be value buys, so would you see any point in doing technical trades on them at this point?

Or better, would you do technical trades, on them? Does your trading style incorporate those types of trades? Perhaps not what you would recommend, but how you would normally play a buy like this?

I figure holding for a year will yield good results, buying and selling might yield some decent results as well, of course I might just miss the entire gravy train if it takes off with constant few % gain per month for the next year or two.

@E-man

I'm actually looking to change my strategy from a mess to something more reliable :) So while we all need our "own" strategy, I find it really important to understand why someone is taking X action and recommending it. I learn best when I understand all the little in's and out's of each trade vs a straight recommendation.

People always bash Cramer, but from reading his books (they were interesting, not great books for investing) and having a slightly better understanding of him, I think he's able to convey some good teaching lessons to his audience. The ones who try and implement his exact buys/sells probably don't do well, but the ones who learn a bit about how to research a stock probably do gain.

Thanks for the update e-man, it helps me also understand where you're coming from! I'm probably aiming for something between the two of you. Doing some technical trading, but mostly value investing. Right now, with the markets gyrating like they are, it seems like just going for a high/lows strategy will pay off. If you miss a cycle, wait it out.

I'm putting about 3K/month into my trading account. Sometimes more. I don't have a 401K program anymore, so I consider it also 401K savings. When the brazilian Rais takes off, I transfer money down there. I currently get 6.5% interest down there. For the last 5 years, the currency has gone from roughly 2.40:1 to 1:80, holding at 1:80 for most of this year. It hit 1.57:1 just before the global melt down, and then spiked to 2.4 for about 2 months, dropping off about 15 cents per month there after, settling at about 1.80 this year. I bought mostly when it was 2.20-1.95.

I was *really* hoping that elections this year would yield a sarah palinesq contender to scare investors off, but they're all been rather bland. The female-gorilla-fighter turned VP has been doing a great job of softening her appearance up, which has reduced the tensions I was hoping for :)

Oh well! 6.5% isn't bad for the near term. I'll either lock it into land, or transfer it back to US currency as it's cyclical spiral out of control, which I suspect will be in 2-4 years.

It won't be true. Rig is calling the shots, not seadrill. (If they say it, they won't do it). I say this from someone who is in the M&A industry and know what it takes to get a deal done. It does highlight the value though.

I would love to see the drilling ban lifted and get the shares back into the 70's. All my friends and family are in as well.

On the Aug. 24 E-man said: I feel that we have one more down day before rebounding to the upside.
Hat tip to E man. I held off a day on my WFC purchase and got in low. I then sweated it out for a week hoping there would not be a total collapse. I've got a stop in now so hopefully it'll be a winner (unless the MM take me to the cleaners).

So far, Rig has been a great winner for me :P While I'm not giving out kisses... lots of thanks are due!

So how would you guys play this one out. I'm eventually looking at getting property in Brazil. Probably for rental purposes right now. Their mortgages are so expensive that prices are kept semi-down. 12% a year keeps + inflation adjusted keeps mortgages down. They're starting to introduce new home buyer packages, which shot pricing way up for any homes within that category. I suspect over the next 10-15 years, things will stabilize and house appreciation will be significant as longer term more affordable mortgages become available to more people. That being said, I have some money in Brazil right now at 6.5% interest. On top of that, the last couple of years have seen 20+% increase in currency value. As stated below, I think the currency is ready to correct within a year or two.

http://www.brazzilmag.com/component/content/article/89-august-2010/12362-brazil-may-offer-tax-cuts-to-encourage-long-term-lending-by-banks.html

This isn't a well written article, but I've read a couple showing debt issues and growth issues that will likely lead to a currency correction. Brazil manufactures most of what it uses via 70% tariffs to protect it's locally produced goods. But with these massive jumps in the currency value, I've started to notice FAR more chinese goods showing up on the shelves. Obviously china is exploring other markets, but the currency had to get to a certain level before they could compete. Now they are. Now Brazilian companies are having a harder time exporting as well. Many products are getting too expensive to export now. So I figure something has to give.

What do you guys think are my best options to protect/invest in this case? For me, I have a couple things going for me. My wife is Brazilian, so we can push through the paper work on buying/investing pretty easily. Father-in-law is banker, we've got at least one real estate agent in the family, and a couple of lawyers in the family. So buying land should be decently easy. Renting out is easy. I rent to someone who has to provider a guarantor, the guarantor has to own 2 pieces of property, it's how they rent down there! Leases are roughly 3 years. If they don't pay, I let the guarantor know. If he doesn't pay me, I simply file a complain and *ALL* of their bank accounts are locked. No more access to their bank accounts, they can't even get a cell phone until this is resolved. It's extremely harsh. Rental units don't include a lot of what we provide. When we rent a place here, we expect a working refrigerator, stove, and perhaps a SINK, but those don't come with places down there. No Sink. wtf.

I can buy land, buy into the brazilian stock market (Maybe buy something solid like petrobras), get a loan at about 12%. I can also transfer down a decent amount of cash to pay off a loan at any time or buy property. End result is to own land down there, with hopefully a couple of rentals.

I'm going to assume the currency is going to go on a slightly wild ride, within 1-5 years. Knowing that, and what I have at my disposal, what tactics could I employee to best maximize a return down there?

If the currency starts tanking, I was thinking I could buy land + get a loan. When it starts to stabilize, shift money from the US down there to pay off the mortgage, hopefully at a far better rate.

I guess I could invest in the stock market down there, buying into something like petrobras, which should negate any currency fluctuations since it's selling product on a world market?

What other tactics might I want to look at?

Oh yeah my 6.5 is compounded every like 3 days. They deposit money into my savings account *all* the time. And it's tax free income down there.

pkennedy says

So how would you guys play this one out. I’m eventually looking at getting property in Brazil. Probably for rental purposes right now. Their mortgages are so expensive that prices are kept semi-down. 12% a year keeps + inflation adjusted keeps mortgages down.

Wow, no wonder there's such horrendous income distribution in Brazil. There's no way for the poor to climb up the social ladder, and there's virtually no way for the rich to fall off a cliff there. The economic system there is clearly rigged to maximize the profits of the cash-rich, and to oppress the poor with high interest rates, and nearly impossible to obtain mortgages for anyone except those with lawyers, bankers, and real estate agents in their family.

The only thing that would make it even better than your serfs(renters) having their cell phones cut off when they don't pay you the rent on time is if they actually still had a debtor's prison. How dare the lower classes become unemployed all of a sudden and choose to feed their children instead of honoring their 3 year indentured servant contract(lease) with their master (landlord). Hell no, cut off their cell phone immediately!!!

The serfs on the left pay rent to those on the right:

Uh no one is going in there to collect rent, to buy or rent anything out. I'm talking about the other side, that is where I would be buying, and renting out. That would be the side with the individual pools.

And the guarantor is the one who gets nailed. He has a minimum of 2 properties and gets everything denied as well. High interest rates are due to inflation, and actually slow the economy down right now.

Social ladder comes from staying in school, and unfortunately that doesn't always happen. They're paying parents to keep their kids in school now. The whole middle class is definitely growing! Florinopolis has an amazing economy, they've taken education seriously and you won't see beggars around. I was fairly surprised, my mother-in-law who is heavily vested in education was explaining all of the programs they've started and the effects they have had on the peoples lives.