Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Perhaps Dell was quite upset about Apple receiving Yonah priority. Hence laptops first, followed in mid-late Q2 with Rev F / 1207 servers?
EDIT (see below) : I've already answered it many times.
Prior to this transaction, he had more invested in AMD, at higher risk & reward, than he does now.
He took a little off the table with this transaction, which makes sense given the run in the stock. He could have taken even more off the table, "cashing out" completely, but he did not. So the transaction, in isolation, was bearish, but not as bearish as it could have been.
Now, to actually evaluate this properly, you need to look at more than this one transaction, as explained. How many shares (stock & options) does he receive & purchase over time, and how often does he sell, like he just did.
Your focus on "direct holdings" is misguided, as I've explained over and over again.
----------------------------------
Another hypothetical which may help:
Position A:
$2M worth of Jan 2008 $10 strike AMD calls.
Position B:
$1.2M worth of AMD shares.
Which position reflects a more bullish attitude about AMD?
Taken in isolation, with this transaction, Palmer moved from position A to position B.
You're very welcome.
Somewhat hidden, but I guess that´s as much of an acknowledgement of my fairly obvious point as I can hope for.
No, it was a refutation of your point.
If the old position was more bullish than the new, than the transition from old to new is net bearish, by itself.
Common sense dictates that one tracks an individual's total position in a given company, not an arbitrary subset of the position that is not necessarily correlated with the total position, if one seeks to derive that individual's changing opinion of that company from the changes in the magnitude of his investment over time.
No, he decreased his exposure to the volatility of the stock rather significantly with this one transaction. His in-the-money options, which he exercised, were more leveraged than the stock he now holds, and thus more volatile, at risk, and a more bullish investment than the stock he now holds.
But again, one really needs to look over time to compare this reduction with grants (or other purchases) he has received, in order to decide whether his overall actions are bullish or bearish.
Quite frankly increasing direct holdings in a certain stock to somewhat impressive levels compared to previous history looks like a bullish sign to me
Well, once again, it ISN'T. Total holdings matter, direct holdings mean nothing by themselves.
no matter what analogies you may come up with to explain something that I told you has been covered many times before and isn´t exactly hard to understand.
You seem not to understand. That's why I'm using analogies to try to help you understand.
Here's another one, repeated:
----------------------
If Ruiz exercised all 4.8M options, sold *everything* EXCEPT for 250K shares, doubling his "direct holdings", you would think this is BULLISH?
It would be hugely BEARISH. Ruiz would be vastly decreasing (indeed almost completely liquidating) his holdings in AMD, even as he doubled his "direct" holdings.
This shows (yet again) why "direct holdings" by themselves (and therefore changes in them) are meaningless.
You (and many others) are essentially analyzing how much cash Palmer keeps in US Dollars only (versus US coins), and using that to decide how he feels about the US currency, rather than looking at the combined total, and observing how that fluctuates with his holdings in Euros.
This is my last post on the subject.
It isn't bullish, bearish or neutral as it stands-- you can't tell without more information. He decreased his TOTAL holdings in the company with this one transaction. You need to know how his holdings have (and will) increase over time to evaluate his net change in total holdings.
I don't know how much clearer I can make it.
Did you see my second edit? Here's the analogy for you:
---------------------
Palmer has $100 in quarters. He goes to a bank and coverts them into 5 $20 bills.
Then he uses 3 of the $20 bills to buy euros.
You think that shows he is bullish on US currency? Why? He now has only $40 total, as far as you know. Instead of $100. Sure, he has 2 $20 US bills, and he had none before. His "direct holdings of $20 bills" increased. Big deal.
You can't tell whether or not he is bullish on US currency without knowing how much US currency he has received in the past, and will receive in the future. And then you can weigh his frequency (and magnitude) of euro exchanges against it.
EDITED(2): It isn't simply "additional information", Keith. My point is that "Direct holdings" is meaningless by itself.
In fact, your Palmer example is not correct. If Palmer recevied less than 36K ( minus some small "diversification" %) worth of stock and options over a reasonable past time period, then it is not bullish, as he is *decreasing* his total investment in AMD.
but don´t act like I don´t know what I´m talking about when I raise a valid point.
In this case, you seem not to. I'd suggest thinking it through again. In fact, I think you owe me an apology.
If it helps, oversimplify, and think of the options as being struck at zero, so they are just "stock by another name". That should make it clear that a decrease of 36K (shares+options) cannot be evaluated without knowing how many shares+options he receives annually, or what have you.
The actual stock sale, by *itself*, is bearish, if you only focus on that transaction, but that is silly. A year is a reasonable time period to assess net flow of insider holdings (both stock and options).
An individual will receive new grants, and they will sell shares. Are they net accumulating to flat, very slowly distributing, or quickly distributing? TOTAL holdings, not just actual directly held shares. That's what matters.
It's like evaluating someone's bank account, looking at checking, and ignoring savings, and thinking you can decide whether their overall account is growing or shrinking. It isn't rational.
Now, everything is made much more complex by the various strike prices of option grants, particularly new grants, vs. old grants which may be exercised and sold.
But the same basic principal applies. You need to look at fluctuations in total insider investments in AMD, not fluctuations solely in directly held shares.
Sell (and exercise & sell) transactions tell you nothing without knowing how many grants an individual was *given* (or purchased with their own cash) over a previous time period.
---------------------------------------
A final analogy for you:
Palmer has $100 in quarters. He goes to a bank and coverts them into 5 $20 bills.
Then he uses 3 of the $20 bills to buy euros.
You think that shows he is bullish on US currency?
You can't tell without knowing how much income he's allowing to accumulate in USD, can you?
Well, once again, those sorts of Yahoo stats are useless, because most insiders have most of their "inside wealth" held in in-the-money options, not shares.
For example:
Yahoo lists *total* AMD insider holdings as 400K shares.
Ruiz has 4.8M in-the-money options.
Need I say more?
Keith, that's the wrong way to look at things. I realize this has been a common sort of post here and on the SI board, but it doesn't make sense to merely look at share holdings.
It is analogous to evaluating a brokerage account (call it the "AMD account") by looking only at cash holdings, and ignoring stocks.
Someone like Ruiz has the VAST majority of his AMD wealth held in exercisable in-the-money options, not shares. We're talking > 90%.
What you want to look at is:
Share holdings PLUS investments held in the form of options (at least in-the-money options)
Now, options with higher strikes don't hold as much value as lower strike options, obviously.
But a good first approximation would be the number of shares backed by options, plus the number of actual shares.
Take Ruiz. You'll want to look at something like this:
How many shares did Ruiz receive in 2005 (say) in the form of direct stock AND in the form of option grants. And only count options that are currently in the money and have value.
And how many shares did he end up selling?
------------------
If you want to attach an interpretation:
Neutral is probably selling slightly more than one receives, as diversification would be expected by any sane individual with most of their wealth in one stock, regardless of their outlook on that stock.
Bearish would be selling much more than one is accumulating.
Bullish would be flat to increasing total holdings. (not just stock)
-------------------
In Ruiz's case, from the 10-Q's it seems likely he'll receive at least 500K shares in the form of options in 2005.
His total sales from 2005 AND this latest Jan 2006 transaction amount to: 300K shares worth.
So he's net positive 200K shares + options.
Bullish.
Now as I said, this is not perfect without analyzing the strikes of any options exercised & sold vs strikes of new option grants.
But focusing only on direct share holdings is not rational, given the amount of worth tied up in options.
For example, if Ruiz exercised the 4.8 MILLION options he holds, and sold 4.4M shares, keeping .4M new shares, that would be incredibly BEARISH, not bullish, despite the large increase in his direct share holdings.
The 10-Q's are a good source of information on this.
BTW, Hector's total AMD holdings should amount to between ~$70M and $130M these days.
He's got 4.8M options in addition to his 200K shares. And the strikes probably average to $20 or lower, meaning he has >$20/option of worth. Could be as high as $26 or so / option. Can't tell for sure.
http://www.sec.gov/Archives/edgar/data/2488/000119312505215504/d10q.htm
An AMD notebook with a USB notebook mouse wouldn't last an hour.
I'm sure you could find one that wouldn't, but the vast majority of new ones would last much longer than that. But I suspect you knew that.
Have you bought in yet? BTW, did you hold your Intel investment over the earnings cliff?
No one can determine with any accuracy what ASPs are except for some massive averaging.
Probably why the "A" in ASP stands for Average, smooth.
You mean, did he ride his INTC investment down to these low levels? Excellent question.
We do know he sold his AMD shares at about 1/2 the level they currently trade at.
It would provide some useful context for weighting his current investment opinions regarding AMD & INTC... whether to give them some small positive weight, or a large negative weight, if they seem anti-corollated with reality.
Not in Q1, they don't.
(And they have nothing exciting in Q2, so that's not going to happen either.)
My argument is that Core-32 notebooks do not perform as advertised (w.r.t. battery power) when a USB device is plugged in, even when the USB device draws essentially no power.
Further, given that the problem has been known (by Microsoft) since July 2005, I submit that it is rather unlikely to be a driver bug, and instead, a bug in Intel's chipset.
------
What you posted in no way proves anything contrary to those assertions. I think the thread has seen about enough on this topic now. You can have the last word.
p.s. bought those AMD puts yet?
LOL! I'm sure folks with a USB flash drive or a USB mouse will appreciate your advice. I guess Core-32 notebooks should only be used within 6' of an outlet?
If it were the driver, it would've been fixed by now, as it is 6 MONTHS after the problem was identified.
Intel makes itself look bad all by itself. It doesn't need any help from me.
If by "elegant" you mean better performance, performance/W, and performance/W/$, with ASPs 2/3's of Intel's, then that's a good portion of it.
AMD's significant capacity boost throughout this year is the other piece.
BTW:
The other reviews didn't bother to test with a USB device plugged in.
Read the opinion post that goes into detail:
-----------------------------
http://www.tgdaily.com/2006/01/29/opinion_core_duo_microsoft_power_drain/
Windows / Core Duo fiasco: An industry asleep at the wheel?
By Aaron McKenna
Published Sunday 29th January 2006 18:54 GMT
Late on Friday evening, after nearly two weeks of testing and investigation, we revealed (http://www.tgdaily.com/2006/01/28/toms_hardware_uncovers_power_drain_issue/) that Intel's mobile dual-core poster boy suffers from a major flaw which sees USB devices sap upwards of 80 minutes of battery life. Intel has pointed the finger squarely at Microsoft, claiming that it's a bug in one of their drivers that's the cause of the power drain.
Apple, who is getting into Core Duo processors in a big way this season had better hope that it is a Microsoft driver which is at fault and not the processor itself. Otherwise they're in as much hot water as the Wintel dynasty is at the moment. As one seasoned journalist put it to me over the weekend, these two (Microsoft and Intel) are such slippery characters that anything could be the cause of the battery-sapping issue. Both companies are so shrouded in mystery, NDA's and a culture of secrecy that it might take us some time to untangle the mess entirely.
We can only imagine that Intel puts enormous amounts of pressure on Microsoft to take the rap for this bug which essentially takes the centerpiece of the Intel's marketing efforts for Core Duo - imporoved battery life - and beats it around the head with a 2x4. However it is strange that this bug only shows up on the Core Duo chipset, and when combined with the likely back-room haggling and arm twisting that went on last week between the two companies in order to see who would take the blame; we have a gut feeling that there is still more to the story than what we know today.
What we know for a fact is that they both knew about the problem since at least 12 July 2005 and with the first Core Duo systems coming to retail this month it didn't look like either party was rushing to tell the buying public about the problem.
Indeed, those who did speak to us about it only did so anonymously and when they were told that we had the goods on Core Duo and were about to start raining on the parade. Others claimed to have no knowledge of what was going down when I approached them about the problem last week.
Instead of throwing out direct accusations and giving specifics I decided to talk to one particular manufacturer, Rock, who had just sent out a well-timed press release on their new Core Duo notebook, the Pegasus 550N. I asked their PR contact, who asked the engineers, if they knew about any issues with Core Duo that had showed up during their testing of the processor or at any other time. They claimed no knowledge whatsoever of what I was talking about. [Update: Rock has issued a "no comment" on 30 January]
This means that either they haven't bothered to rigorously test the part and were not privy to internal Microsoft or Intel rumination's; or they were keeping mum, presuming that I was a journalist who had heard a few rumours and wanted to scare something interesting out of them.
I would of course point out that they deserve the benefit of the doubt to an extent; in that they simply might not be able to notice a strong hint if it landed on their shoulders and started to make disparaging comments about their mothers.
Any of these three possible explanations to their lack of disclosure on the subject doesn't paint them in a particularly flattering light, but it must be said that they were stitched up: I didn't tell them exactly what I knew and I didn't give them the chance to anonymously come clean and denounce Wintel to us without having to worry about a backlash from the all-powerful dynasty, as their peers have done.
All of this combined - the silence of Wintel and the notebook manufacturers, as well as the palpable lack of a fix for the seemingly simple problem - doesn't paint anyone in a particularly flattering light.
Certainly the lack of the fix to what is supposedly a simple registry error will give license to conspiracy theorists to claim it is a problem with the chip itself and not Windows XP SP2. This would certainly turn quite a few heads towards Apple, whose partnership with Intel is starting to look a little less rosy than it was a week ago.
What really and truly ticks me off is the fact that between them Intel, Microsoft and the system manufacturers have kept their mouths shut for six months and were going to start selling these machines without acknowledging the bug or providing an immediate fix for it. The sad fact is that I think they would have got away with it, too.
The failure of the technology press
I would say that the calculation was made that in the first month or two they would be able to sell thousands of Core Duo machines and not have to worry about the problem cropping up until users began digging deeper. This is a massive, simply incredible, failure of the technology press.
I've read scores of Core Duo reviews and previews over the past month. After being told of this particular problem a week ago I began to read them more closely, and I've turned up an impressive array of quotes for my personal scrap book of infamy: "While offering 14.5% longer battery life, the Core Duo notebook also brought us a 16.6% increase in performance. More performance and longer battery life? Who could ask for anything more?" was the quote from one review.
There were plenty more like that, all of which begs the question: Why is it that it was only Tom's Hardware Guide reviewers who turned up this problem? Judging by the problem, stemming from USB device use as it does, the reason we believe we found the problem is not just our enthusiasm for what we do, but also the diligence and many hours of research we put in every product that crosses our desks. Putting in that extra bit of effort seems to have paid off this week.
Maybe everyone just had a bad day with their Core Duo tests (may they grant me the same benefit of the doubt on the day I put a stone through the glasshouse window), but I think the problem is laziness. Most reviewers probably thought to themselves "Stick in the benchmarks, make sure you have a clean install of Window's for each one but let's not do more than we bloody well have to..."
Testing takes a long while (ask anyone who has ever had to install SysMark onto a machine...) and can be quite boring and mundane at times, particularly when you're doing it several times a week to several systems at once. So you can see where the mentality comes from to avoid doing more than you really have to. The problem with this is that as the watchdogs we're supposed to do the rigorous testing and research, or else people will wind up spending their hard earned money on products that aren't what they seem to be.
This is not a "We're Holier-Than-Thou" rant, and we're not looking to laud our victory and keep score. Indeed, far from it. I don't want Tom's Hardware Guide to be in any way, shape or form "Holier-Than-Thou" with regards to its competition.
This is because I am a consumer as well. I want to be able to read a variety of reviews of a product and know that each and every one I read has been written by reviewers who are all equally as thorough as the other.
I don't want to have to do as one has to with the current crop of Core Duo reviews and keep my eyes peeled for that one review that notices the "minor" flaw that sees my soon to be new notebook cutting its battery life by 80 minutes because I've had the audacity to plug something into the USB port.
Sadly, this probably isn't the first or the last time we'll ever see the likes of Intel and Microsoft keeping something from the public for their own good. But the press shouldn't be coming out of this as badly mauled as the people over whom we are supposed to watch. A lot of reviews will now be cast into doubt as readers ask themselves "Well, if they couldn't find the flaw with Core Duo what else are they missing?"
We review products so that you know what you are getting when you spend your money. This is the unwritten agreement that you sign with us: You provide us with page impressions which we use to sell advertising to keep ourselves in business, and in return we give you the unbiased and thorough low-down on everything that crosses our desks. Nobody in the press is guilty of corruption in the Core Duo case, but they are guilty of complacency and negligence.
They need to wake up and smell the coffee. One among us shouldn't have to be the watchdog who puts in longer hours of testing in order to make sure as sure that a product will not disappoint and anger consumers. Sure we'll get plenty of praise for having been the one reviewer among many to have caught out Intel and Microsoft on this one, but as a consumer and a reader of many reviews I don't want to have to rely on just the one publication.
It follows from their respective revenue guidance for Q1. Remember Bryant's Q4 comment at the December update? You don't think he feels the same way about Q1, given a similar guidance setup?
Intel midpoint is -8%.
Historically, AMD uses "slightly down" to mean -3% or so, so their midpoint is -1.5%.
Unless you imagine some weird allocation, in which Intel's non-CPG areas are responsible for the -8%, AMD is going to gain more share in Q1.
And Intel knows it.
If Intel sold cars... "Well, sir, it's true that your new car doesn't get the MPG we told you it would, but it's better than the old SUV you traded in for it, so what're you complaining about?!"
Intel is expecting further AMD marketshare gains in Q1, right?
I guess that makes Intel bullish on AMD stock, as it pretty much follows from both firms guidance for Q1, combined with their product offerings in H106.
I don't see them making $0.50 per share this quarter.
You didn't see them making $0.45 on 1.3B in revenue in Q4, either.
Besides SPSN loss percentage declining from 60% to 38% (if you care about non-cash charges) consider that the bonus schedule was likely reworked, so that *cash* expense, now hidden inside CPG op inc, is also likely to decline.
Yes, the point is that *any* connected USB device seems to trigger bad power behavior.
That's a novel standard you and Alan seem to have regarding new Intel products.
Performing as promised isn't required. Rather, being "at least as good" as some previous part is all that matters.
I suspect Intel's customers would beg to differ.
We now have 2 more examples of execution troubles:
(1) Broken EIST on the first batch of Preslers
(2) An apparently faulty Core-32 chipset, resulting in bad power handling when any USB device is attached to a notebook.
and they are still hitting record revenues and unit sales.
The market is forward-looking. Intel is forecasting a YoY revenue *decline* in Q1, which means they will lose more marketshare to AMD, given AMD's guidance.
How do you see a forward P/E of 20 as an "absurd level", given the projected rate of earnings growth? I doubt many would agree.
I think the datapoint is still interesting, even if it is based on "here's where it would fall on the currently published list, based on its performance." I'm confident that most readers are aware that faster systems are released all the time, and can take that into account.
The question is, were those reviews running with the broken EIST turned on in Presler, or not? If you turn it on, in B-1, you risk a complete system hang, due to the problems fixed in C-1. But did the reviews publishing idle power have it active, or not?
Keith, did you see the tech report story? Or are you saying that idle/sleep power of the broken Preslers is still better than Smithfield?
http://techreport.com/onearticle.x/9256
Akiba PC Hotline! is admonishing prospective buyers that the first batch of Pentium® D 900 series of processors that have arrived in Japan may not have Enhanced HALT State and Enhanced Intel SpeedStep® Technology (EIST) enabled. Apparently, one shop is saying that because EIST and other power-saving features are not fully functional with this initial lot of processors, the TDP (thermal design power) might go up as much as 10 - 15W. It should be noted that Intel's Pentium® D 840 has a TDP of 130W, but the 920 and 930 have lowered that to 95W. The 940, 950, and XE 955 are rated at 130W which is the same as the Pentium® D 840 (see table 5-1: processor thermal specifications on page 76 of the datasheet for the Pentium® D 900 series here.)
Akiba goes on to highlight a footnote to the Intel Pentium® D Processor 900 sequence and Intel Pentium® Processor Extreme Edition 955 datasheet and specification update stating that Enhanced HALT State and EIST will be functional for specified units of these processors through a BIOS update which is expected some time in the second quarter of this year. However, what may be particularly troublesome is that Intel's latest errata (AA30 on page 24) says that during an Enhanced HALT or EIST ratio transition, the system may hang. Indeed, it is hoped that future BIOS updates alone can resolve these issues.
Whatever the case may be, this revelation, if true, could help explain some of the clock throttling issues that Dr. Damage encountered during his testing of Presler which he wrote about in this post.
http://www.techreport.com/onearticle.x/9198
Keith, "theflyonthewall" is largely a rumor site, so it's a perfect match.
Appro 1024-core Opteron machine makes #132 on Top-500 supercomputer list
http://www.theinquirer.net/?article=29324
Yonah (Core-32) power troubles with Windows:
945M chipset to blame?
http://www.tgdaily.com/2006/01/28/toms_hardware_uncovers_power_drain_issue/
http://www.theinquirer.net/?article=29326
[...]
But that's debatable - according to the site. It claims that the Napa chipset is the real problem, and quotes a number of unnamed notebook manufacturers saying the 945GM chipset is the real botheration.
May want to rethink that AMD put purchase:
http://www.siliconinvestor.com/readmsg.aspx?msgid=22110933
I guess stepping B-1 (the current Presler) stands for "broken". All the suckers who bought (or buy) one of the those are stuck with a little furnace. C-1 is the stepping they should've delayed the launch for.
You appear to be confusing server parts & sockets with the desktop socket.
No. They aren't.
Rev F is the next version of the K8.
M2 is the desktop socket.
S1 is the mobile socket.
1207/"Socket F" is the server/workstation socket.
It wasn't a last minute delay. It's been known for months, just not by Charlie, apparently.
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2584
November 1, 2005
Our initial roadmaps claimed MCP55 would show up right around Cebit 2006, which made sense considering we had also heard Socket M2 would launch then. However, these newest NVIDIA roadmaps claim that MCP55 won't show up on motherboards until late Q2'06; right in the June/July timeframe. Unless AMD is planning to launch Socket M2 on a VIA chipset exclusively for three months (I'll give you a hint, they won't) don't expect AMD DDR2 processors to launch at Cebit anymore. Low end C51PV and MCP51 chipsets will also launch right around the same timeframe specifically for Socket M2 Sempron.
------------
Nothing to do with Fab 36, or the Rev F parts themselves. It's waiting on nVidia's desktop chipsets. And it's been known since October.