AMD Q3 Results

[AtenRa]

Didn't listen to to the call? AMD predicted a $200M - $250M loss for 2014. Quite a ways away from your $165 profit.

This is operating Income, not profit. They lost more in 2013 than they are going to loose in 2014. So, 2014 is better than last year, unlike the doom mrmt is trying to promote.
Don't get me wrong, it is not the best they could do but it is better than last year. mrmt would doom everything AMD even if the make 1B profit.

 

[mrmt]

For 2014, they are projecting cash flow positive. Are you saying that from the call, you heard they're projecting to have a $375M-$425M loss in Q4 on 1.243B in revenue? The company would be bankrupt today if that were the case.

Here's Kumar's quote on the matter:

http://seekingalpha.com/article/256...-results-earnings-call-transcript?part=single

(...)

And finally, free cash flow is expected to be positive in the fourth quarter; and for the full year 2014, negative in the range of $200 million to $250 million, including the impact of the $200 million special payment to GlobalFoundries earlier this year.

 

[hungtran]

Here's Kumar's quote on the matter:

Well, at least that $200M was the final payment for the 2012 WSA penalty and should not be factored into 2014 cash flow performance. Now it's breakeven to $50M loss, but includes the extraordinary cost to redeem and rotate their debt forward. With the poor performance since, it really looks like they got a good deal on those bond issues.

 

[mrmt]

Well, at least that $200M was the final payment for the 2012 WSA penalty and should not be factored into 2014 cash flow performance. Now it's breakeven to $50M loss, but includes the extraordinary cost to redeem and rotate their debt forward. With the poor performance since, it really looks like they got a good deal on those bond issues.

AMD results breakdown is like that:

Gross revenue....... 1.429MM
Gross Margin......... 35%
Gross profit........... 494MM

R&D..................... 278MM
SG&A................... 150MM

interest expenses.. 46MM

(I'm using 46MM for interests since this is the Q4 forecast)

AMD basically needs 474MM to cover its basic expenses, R&D, SG&A and interests. Today it basically generates a few million here or loses a few millions there, it's more or less stable. But what happens if they take a 13% drop in revenue, 1.243MM with 35% gross margins?

Gross revenue....... 1.243MM
Gross margin........ 35%
Gross profits......... 435MM

AMD only has 435MM to cover its basic needs, but it needs 474MM. They go from generating small sums of cash to burning cash, 40MM per quarter, 160MM per year. Add Q1 and Q2 seasonality and 200MM in cash outflows seems reasonable. This is why the results were bad, quarterly results were fine but the forecast is anything but good.

But since AMD is cutting 7% of its workforce, saving about 21MM per quarter, that gap shrinks a little to roughly 20MM per quarter, but again we should add the traditional seasonality in Q1 and Q2. Given that AMD has nothing good on the 2015 pipeline, it's safe to say that 2015 will be a very long year for AMD. They have to chose between firing more personnel (another 10-15%) or will burn precious cash to survive in 2015.

But there's an interesting bit about the latest round of cuts. AMD is firing about 7% of its workforce, if 100% of it is SG&A then it's about 13% of its SG&A force. If marketing and sales bear the brunt of the cuts, we should be talking about cutting above 25% of its sales and marketing force. I can't think of cuts of this order or magnitude if the executive management wasn't planning to shift the company to other markets that demand smaller sales effort or if management wasn't forecasting sales to stay at a much lower levels than in previous quarters.

 

[Phynaz]

Well, at least that $200M was the final payment for the 2012 WSA penalty and should not be factored into 2014 cash flow performance.

Of course it should be factored in. It's cash they had to withdraw from the bank and pay out.

 

[krumme]

As they said they werent going for the smaller design wins. So yes they will have to alter their markets to an even further degree. Makes sense with the transistion they are still in.
We had this talk at the Lisa thread and i am surprised the cuts isnt deeper both 3 years ago and now. So my guess is the same as yours; more will follow 2015 at the same order at least. UNFORTUNATELY.

 

[Phynaz]

This is operating Income, not profit. They lost more in 2013 than they are going to loose in 2014. So, 2014 is better than last year, unlike the doom mrmt is trying to promote.
Don't get me wrong, it is not the best they could do but it is better than last year. mrmt would doom everything AMD even if the make 1B profit.

Operating income is a completely weaselly metric. It's "We would have made this amount of money if we didn't actually have to pay the bills".

Free cash flow is what pays the bills and net income is what goes in your pocket and gets invested in the future.

That said, I believe even operating income is going to be negative this year. I don't think you understand how ugly Q4 is going to be with the predicted sales drop.

 

[hungtran]

The retail low cost laptop market is doing bad. They're losing to bay trail in the $250-$400 category and Su said she won't chase low margins. The recent price reduction on the FX chips and GPU means demand is low there. The revenue to R&D spend ratio could change if there're more semi-custom wins because R&D expenses are shared with customers. I'm guessing there's interest for AMD semi-custom work but they need to pick opportunities wisely because of limited resources. I think they need to take what they can get to improve market confidence. The low stock price and negative analyst views could scare potential semi-custom customers away.

 

[krumme]

Operating income is a completely weaselly metric. It's "We would have made this amount of money if we didn't actually have to pay the bills".

Free cash flow is what pays the bills and net income is what goes in your pocket and gets invested in the future.

That said, I believe even operating income is going to be negative this year. I don't think you understand how ugly Q4 is going to be with the predicted sales drop.

If you want to track the core business operating income is vital. Imo especially regarding a company like amd i would monitor the operating income closely as a better key figure for health than net result. But yeaa all the big numbers can all tell bs.

 

[hungtran]

Operating income is a completely weaselly metric. It's "We would have made this amount of money if we didn't actually have to pay the bills".

Free cash flow is what pays the bills and net income is what goes in your pocket and gets invested in the future.

That said, I believe even operating income is going to be negative this year. I don't think you understand how ugly Q4 is going to be with the predicted sales drop.

Well, the Q4 prediction is already out there, so unless AMD is hiding opex and inventory issues, the operating income is predicable. Whether you believe the prediction is optimistic or sandbagging, that's another thing. My guess is in Q3, Rory underperformed his sandbagged figures, which lead to his sacking. Anyways, there'll be a lot of severance payments in Q4 that will also hurt cash flow.

 

[krumme]

RR just didnt understand what he had in his hands. Did anyone ever get the feeling he was in touch?
Placing him in front of spitzenklasse engis was just a weird theater from day one. But somehow it got along.
Your sandbagging sounds very plausible but its the oldest trick in town and imo thinking about it just couldnt be used as other than an excuse.

 

[krumme]

I have trouble acepting the talk about only sg&a cuts. Looks like investor bs to me.
There is no way they can know what technical competences they need for future embedded projects.
And how is the x86 big core going? What about all the people on excavator project. Can they be moved?

Clearly eg marketing and sales is still in for a huge overhaul as amd transforms each day. As evident from the financial results. But imo some adaption on technical side must also be needed. Cust but also investments.

 

[Phynaz]

Well, the Q4 prediction is already out there, so unless AMD is hiding opex and inventory issues, the operating income is predicable. Whether you believe the prediction is optimistic or sandbagging, that's another thing. My guess is in Q3, Rory underperformed his sandbagged figures, which lead to his sacking. Anyways, there'll be a lot of severance payments in Q4 that will also hurt cash flow.

AMD has missed under in what, the last four quarters? Is there any evidence they will break with their recent history? It's the same people running the operation.

 

[mrmt]

I have trouble acepting the talk about only sg&a cuts. Looks like investor bs to me.

I don't. They may cut the unnecessary SG&A now, but cut down the R&D people once the desired product reach the market.

 

[positivedoppler]

You know, I can't help but to wonder was there a fundamental difference in R&D allocation that lead to the replacement of Read by Lisa so suddenly. Thinking back, a few things comes to mind. Maybe that's why Lisa kept echoing I want to build great products and I don't want to cut R&D.

1. AMD rode the R2 video train for too long. When 20 nm was delay, Nvidia seem to have reacted quickly and cough up the R&D to build Maxwell to 28 nm. AMD did nothing and they might suffer as a result by leaving a few hundred million on the table in the 4th quarter.

2. The FX line is long overdo for a refresh. There should have been a 28 nm 8 core version. If not, there should have been a Carrizo version on the roadmap for 2015 if the performance boost and power saving was as good as advertised. By not investing in the R&D for that, AMD might have left another few hundred million on the table. Intel has not made much performance gain the past two years and left the door wide open for AMD. AMD chose not to spend the money to walk in. They don't have to win the performance crown, they just needed a product that's good enough to gain some market share. An updated FX that's even 20% faster at 30% less power for 2015 can gain significant sale.


Stating that Bulldozer was not the game changer was a gut blow to probably a lot of AMD fans. I don't think Read should have said that in public because not only does that make the FX line even less appealing(which they are still trying to sale), it takes some of the wind out of Kaveri.

 

[VirtualLarry]

Y2. The FX line is long overdo for a refresh. There should have been a 28 nm 8 core version. If not, there should have been a Carrizo version on the roadmap for 2015 if the performance boost and power saving was as good as advertised. By not investing in the R&D for that, AMD might have left another few hundred million on the table. Intel has not made much performance gain the past two years and left the door wide open for AMD. AMD chose not to spend the money to walk in. They don't have to win the performance crown, they just needed a product that's good enough to gain some market share. An updated FX that's even 20% faster at 30% less power for 2015 can gain significant sale.

I agree. I mean, the new FX SKUs are nice, but something with a newer process, better power consumption, and performance, would always be welcome.

 

[NostaSeronx]

The FX line is long overdo for a refresh. There should have been a 28 nm 8 core version.

The FX line lost focus when Barcelona was cancelled, which meant no Phenom FX. Instead, the HEDT just got Phenom which was Budapest/Perseus.

FX was going to get a refresh in 2012, but it would still be the three platform strategy.

G2012 - Terramar/Dublin
C2012 - Sepang/Macau
What would become FM2+ - Komodo

The next FX refresh is planned in 2015 with Excavator/16-core XV.

---
There is also an issue with Excavator and FP256.

FP256 = One macro-op(Single macro-op) for AVX256.
FP128 = Two macro-ops(Double macro-op) for AVX256.

Bulldozer through Steamroller; Those designs can only decode two AVX256 per decode group.
If Excavator supports FP256. Then, the design can decode four AVX256 per decode group.

The decode has to be overhauled again.
The retire has to be overhauled.
The FPU has to support 2x the instruction width, complete redesign.
The LD/ST has to support 256-bit reads and writes, etc.

This is if Excavator is indeed supporting FP256. Essentially, Excavator to support FP256, it must be a new micro-architecture.

 

[AtenRa]

Operating income is a completely weaselly metric. It's "We would have made this amount of money if we didn't actually have to pay the bills".

Free cash flow is what pays the bills and net income is what goes in your pocket and gets invested in the future.

That said, I believe even operating income is going to be negative this year. I don't think you understand how ugly Q4 is going to be with the predicted sales drop.

Ok, lets see Net Income

Q1 2013 = -146M
Q2 2013 = -74M
Q3 2013 = +48M
Q4 2013 = +89M

Total = -83M

Q1 2014 = -20M
Q2 2014 = -36M
Q3 2014 = +17M
Q4 2014 = -35M (estimation)

Total = -74M (estimation)

So again, they are not doing worst than last year.

 

[Phynaz]

Seriously? You going to pull numbers out of the air and use them as fact??

I think I'm done here. There's no point.

 

[MiddleOfTheRoad]

Seriously? You going to pull numbers out of the air and use them as fact??

I think I'm done here. There's no point.

That's because you have no where to go. AMD had an absolutely dismal first half of 2013 -- so there is no way to argue that 2014 isn't at least going to be at least a modest improvement by comparison.

 

[AtenRa]

Seriously? You going to pull numbers out of the air and use them as fact??

I think I'm done here. There's no point.

WTF are you talking about ??? those numbers are taken from AMDs 2013 and 2014 earnings calls.

Have a look yourself in the link bellow.

http://ir.amd.com/phoenix.zhtml?c=74093&p=quarterlyearnings

Just because numbers doesn't suit your agenda doesn't mean they are pulled from thin air.

Next time do your home work

 

[Ramses]

Hey hey hey hey now, no real world actual facts in an internet discussion.

 

[Phynaz]

WTF are you talking about ??? those numbers are taken from AMDs 2013 and 2014 earnings calls.

Have a look yourself in the link bellow.

http://ir.amd.com/phoenix.zhtml?c=74093&p=quarterlyearnings

Just because numbers doesn't suit your agenda doesn't mean they are pulled from thin air.

Next time do your home work

Show me their estimate of a $35M net loss for Q4 2014. Facts only please.

 

[AtenRa]

Show me their estimate of a $35M net loss for Q4 2014. Facts only please.

http://forums.anandtech.com/showpost.php?p=36826897&postcount=129

mrmt estimated 40M, I estimated 35M. Even with 40M loss they will still be better than last year overall.

So again, they are not doing worst than 2013. They are not doing better either but mrmt always make it look bad no mater what

 

[Haserath]

http://forums.anandtech.com/showpost.php?p=36826897&postcount=129

mrmt estimated 40M, I estimated 35M. Even with 40M loss they will still be better than last year overall.

So again, they are not doing worst than 2013. They are not doing better either but mrmt always make it look bad no mater what

It looks more like it's going to be double those estimates but we'll see.