just when amd decided to trade 50% of its revenue stream for midget margin x86 custom chips, and surrender all higher margin pc, server and laptop cpu market share to haswell and nay trail, the abandoned market rebounds.
say, now that all the ps4 launches for the quarter are done, and sony is just at 2.1 million units, can sony make up the promised additional 2.9 million units before march? with ps4 outselling xbox one 2-1, what happens to amd when console sales stall in q2 and amd is left with a 50% single quarter revenue reduction?
the fair value of this pos is $1.50. and that's exactly where it will head after q4 turns out not to be what amd management promised.
i think he's right.
in the q3 cc transcript kumar says "In the third quarter, we made progress towards our goal of diversifying our product portfolio, as we successfully ramped our semi-custom products. Specifically, our semi-custom and embedded revenue accounted for more than 30% of total company revenue, exceeding the target of 20% of revenue by the fourth quarter of 2013."
total revenues in q3 were $1.46 billion. 30% is $438 million in revenue.
kumar also spilled the beans during the cc that margins in console chips were in the "mid-teens". previously in q2 guidance had been "low double digits". of course, low double digits to most people would mean something like 30%. but not to amd management who routinely minces words on the ccs and try to conceal bad data as much as they possibly can.
if mid-teens means ~15%, that means that on the $438 million in console revenue, amd's income was a mere $66 million. that's truly trivial, really, and borderline irresponsible from a management perspective.
read shared that in q3: "Most significantly, our semi-custom business ramped in the quarter. We successfully shipped millions of units to support Sony and Microsoft, as they prepared to launch their next-generation game consoles."
given amd's propensity for being inspecific with key financial figures in ccs, that could mean anything from 2 million to 10 million, but if we take the street's assumption that amd sells sony and microsoft the huge chip for around $100, we see that amd likely sold 4.3 million console apus in q3. total demand for q4 according to sony is 3 million ps4s worldwide. sony leads micorsoft in demand according to most commentators, so at best microsoft will sell slightly less than 3 million xbox ones for q4. if total maximum q4 shipments of consoles is around 6 million units worldwide (in line with sony's estimates), amd did sell most of the chips needed for q4 during q3.
david wong, wells fargo: "One final one, I think you noted your semi-custom business operating margin was in the MID-TEENS, so that was quite a bit better than you’d expected initially. Can you improve it from here, or is this the right operating margin for us to think of for the current consoles semi-custom business?"
devinder kumar, amd's cfo: "I mean the operating margin, when we discussed this last quarter, I had said low double digits, it came in at the mid-teens, and really that’s credit to the higher revenue that we had in the quarter and the successful execution of a very steep ramp in the product in the fiscal quarter production."
got that? console chip margins are in the mid-teens while company gross margins are 35-36%.
call it 15% to be charitable, though it's probably closer to 13-14%, knowing how amd management often minces words and dances around the truth.
15% margins in console apus. that's a friggen joke. that's akin to the margins hardware manufacturers make on nails and screws.
that means that if amd is getting $90-100 a chip for these, that's $15 or less in income per chip. what a totally ridiculous line of business to waste the company's time pursuing. when will the shareholders sue for mismanagement. when you're in debt up to your earholes, you don't mess with breakeven business!
first, check the volume. amd is being largely ignored. short covering would require high volume.
second, nice selfcornhole! you're good at it.
Will this trigger mass returns on the delamination prone R9 290X cards purchased over the last two weeks? Looks like AMD has big problems on its hands.
And I thought that I knew self-cornholin'!
famous last words.
the faces change, but the retardation remains at amd.
amd's biggest problem now just like back in the K5 and K6 days isn't volume. it's margin. every time amd elects to go up against better-funded competition with a price competition, low margin high volume play, it gets killed.
amd has some of the worst margins in the semiconductor industry. it therefore has the least leeway on price. by starting a price war with a guy who can go lower than you can and still make more money than you do, you're asking for a beatin'.
amd is about to get yet another beatin'. this time from nvdia. usually it's intel giving the beatin'.
watch and learn fantards. watch and learn. trust me - you won't make it up on volume.
you should be worried jj. remember when you said that tehmash was gonna see lots of tablet design wins? wrong! remember when you said that richland was going to revive amd laptop sales? wrong again! how many times do you have to be wrong before you realize that you know nothing about the market for cpus?
i know selfcornholin and believe me, you are a selfcornholin master.
i wouldn't rush. but if you do, be sure to be ready to short again the day of the q4 earnings release. you have a 60% chance of winning big on such a position. those are far better odds than you'll ever get in vegas. i covered my $4.10 q3 short at $3.28 the day after earnings, but haven't yet bought any shares to cash out on the pre-earnings runup for q4. much like intel's processor development cycle, i employ a tick-tock trading strategy for amd. it rarely lets me down and over the last 8 years i've made more than $4 million trading amd in this manner.
did you write that article? it contains a sufficient number of erroneous statements that you could have written it.
let's look at a few of them.
"You see, Apple has made a series of interesting moves in the graphics arena lately.
First, a few weeks ago, the Apple announced that the entry-level version of its new 21.5-inch iMac will feature integrated Iris Pro 5200 graphics chips from Intel (NASDAQ: INTC ) . To be sure, that's a solid win for Intel, especially considering NVIDIA scored all of the coveted spots in Apple's iMac lineup last year."
why would that be interesting? for the first time intel's integrated graphics achieves mid-range performance, and in fact blows away amd's quad core apu integrated graphics for all but the very top of the line richland apu. why wouldn't apple put a lower cost, satisfactory solution in its entry level imac? it cut imac prices as part of this refresh after all.
"To NVIDIA's credit, however, and much to Intel's chagrin, it's worth noting each of the higher-end iMac models will still feature various versions of NVIDIA's GeForce 700 series GPUs."
also an idiotic statement. why should intel care that the higher end imacs use discrete nvidia gpus? they still use the family of intel cpus as the entry level model, albeit faster much more costly models, so intel makes more money per unit higher up the imac line. its a win-win for nvda and intc. the only loser here is amd who used to have this business but lost it to nvda when apple moved to sandy bridge cpus.
as intc continues to extend its overwhelming dominance in cpu performance over amd, and to improve the performance of its integrated gpus to levels unmatched by all but the very top of the line currently available apus, nvda is the beneficiary. as intc kills off amd, nvda owns high end graphics where the margins are.
smart move. it was a stupid idea. a low margin, limited scale market, with no barriers to entry and facing overwhelming competition from a host of internet content deliverers. internet tv projects are almost as stupid as supplying chips for game consoles in the face of mobile gaming which is growing annually at the rate of 41% (compared to console gaming which is projected by gartner to be growing at just 16% over the next year).
actually it turns out that the problem with the website was that verizon implemented a couple of freebie opteron servers given to it by amd as sweeteners for the recent xeon microserver deal so that amd could truthfully claim that verizon was receiving both opteron and xeon servers in the deal. sadly, when verizon deployed these opteron boxes in its environment, it totally bogged the entire network down and tripled the datacenter's power consumption.
how could they do that, ace? amd is selling sony and msft the same kind of x86 apus it's unable to sell in laptops and pcs. amd doesn't make anything. it has tsmc make the chips for it. so long as tsmc doesn't blow it, there should be no problem fabbing all the two-generations-behind apus the console makers want.
correction - "2 or 3 spewing lying trolling maggot moles" who told you repeatedly that it was going to drop on the q 3 results, and who tried like all get out to save you from a 25% haircut, but you were too stupid to listen.
i've been actively trading amd since the mid 2000s. it is almost impossible to play it wrong. you sell shares you've held since the post-results selloff, short on the eve of the current quarter's earnings just as #$%$ like you are getting suckered in and pushing the price up, cover shortly after earnings, buy some shares to ride the hype and speculation train back up, and then sell and short again. wash, rinse, repeat.
amd has always been a trading stock. now more than ever now that it is a small cap and 20% of the float/115 million shares are short. only a complete know-nothing moron would buy and hold this perennial loser. here's a spin on einstein: stupidity is doing the same thing over and over again and expecting different results. my own trading philosophy with respect to amd parrots another famous saying: let us be thankful for the fools. but for them the rest of us could not succeed. quarter after quarter, year after year i make money trading this stock while numbskulls like you make stupid predictions and lose your shirts.
do you really know anything about amd price action? anything at all? merely meeting expectations will signal a huge selloff. slightly exceeding expectations always triggers a selloff. amd hypes, whispers, and pumps expectations so aggressively back-channel, that not even the whisper number comes close to the real expectation among analysts rating it as a buy. every single quarter this smoke blown up analyst fans' butts dissipates, triggering a big selloff.
to go up, amd must exceed expectations and provide better than expected guidance by an enormous margin (200%+).
don't believe me? then do your own due diligence:
"Over the long term, AMD favors a narrowing pattern, either cutting back or reversing its evening trade 19 times in the last 34 quarters, and seeing flat trade between the sessions in another instance."
that seeking alpha article provides 19 data points dating from 2013 back to 2004. whenever amd matches expectations or beats slightly, it gets pummeled. that's a well-know fact that anyone trading amd should know very well.
i think you meant "finally". but look on the bright side. your banjo playing is much better than your english.
raymond james? who's that? never heard of them.
goldman, bernstein, b of a, merril, credit suisse, citi, and other leading analysts all say that amd is "scrooed".