AMD Q2 2013 Market Share up 2.2% Q to Q (Mercury Research)

[AtenRa]

Not bad, Q3 could be even better.

AMD's share was up in Server (+0.3%) Desktop (+1.7%) and Mobile (+2.9%) over Q1.

http://investorvillage.com/mbthread.asp?mb=476&tid=12970932

 

[SiliconWars]

Good, but they must keep improving to get back to where they were before it all went pear-shaped last year while keeping opex down near current levels.

 

[inf64]

Looks good but like SiliconWars said they have to continue improving like this.

 

[sontin]

They shipped Kabini/Temash and Richland.
On the other hand their gross margins are down to 39% from 46% last year...

 

[Vesku]

Not a surprise on the gross margins, lower price is one of the main ways to gain marketshare even more so in a weak market. Less units sold this quarter than same quarter last year. Wonder how many of those are Jaguar chips?

 

[raghu78]

AMD's biggest problem is their big core execution. the steamroller based kaveri is delayed to feb 2014. there are no steamroller based opterons for 2014 which is frankly weird.

http://www.amd.com/us/press-releases/Pages/amd-unveils-2013june18.aspx

they are going for Berlin APUs for server market which is not a good idea for traditional server market which needs integer performance. but the biggest headache seems to be globalfoundries who are screwing up big time on the manufacturing side. 28nm high performance seems to be the cause for kaveri delays (probably yield issues).

also with Baytrail Intel will hurt AMD badly. thats the last segment in the market where AMD was selling well. Now even that will be under severe pressure.

AMD's CPU division has to do a much better job at executing and that means their foundry partners also need to step up.

 

[Vesku]

AMD's marketshare is so low in servers that they can make a lot of gains even by addressing niches such as HPC, Berlin APU + GPGPU, and dense clusters, Seamicro acquisition + Cat family x86/64-bit ARM SoC. If those gain them just ~5-6% more over all server share then they've more than doubled their market presence.

AMD just does not have the management nor resources to compete head to head with Intel in the "traditional server market" and probably not in any market that doesn't value their GPU advantage or their willingness to offer lower priced product. Intel is in a different weight class than the rest of the competition due to maintaining the lead in fabrication. This won't change until Intel hits a fab wall that they can't spend their way past. There are already some signs that Intel may be slowing down in that regard due to market pressures (less Capex to power through on the fab side) and the increased technical difficulties of continuing to shrink transistors: http://www.reuters.com/article/2013/07/17/intel-results-idUSL1N0FN20G20130717

 

[littlebluesmurf]

Still down of where they were a year ago, hopefully they can keep the same momentum in the next quarter.

 

[raghu78]

AMD's marketshare is so low in servers that they can make a lot of gains even by addressing niches such as HPC, Berlin APU + GPGPU, and dense clusters, Seamicro acquisition + Cat family x86/64-bit ARM SoC. If those gain them just ~5-6% more over all server share then they've more than doubled their market presence.

AMD just does not have the management nor resources to compete head to head with Intel in the "traditional server market" and probably not in any market that doesn't value their GPU advantage or their willingness to offer lower priced product. Intel is in a different weight class than the rest of the competition due to maintaining the lead in fabrication. This won't change until Intel hits a fab wall that they can't spend their way past. There are already some signs that Intel may be slowing down in that regard due to market pressures (less Capex to power through on the fab side) and the increased technical difficulties of continuing to shrink transistors: http://www.reuters.com/article/2013/07/17/intel-results-idUSL1N0FN20G20130717

AMD need to find a way to be competitive with Intel big cores. Its not impossible. With Athlon K7 and Athlon 64 AMD showed its possible. Jim Keller the architect of Athlon 64 is back and his leadership results could be evident by late 2015. Also AMD need their fab partners to deliver competitive process tech compared to Intel for high performance CPUs That is the bigger problem because the foundries are all bothered about low power process since mobile market drives demand for foundries. Intel dominates the server market because of excellent microarchitecture and also world class manufacturing process. AMD has to get Globalfoundries to work with IBM on FDSOI for high performance CPUs on future nodes.

 

[jpiniero]

AMD need to find a way to be competitive with Intel big cores. Its not impossible. With Athlon K7 and Athlon 64 AMD showed its possible. Jim Keller the architect of Athlon 64 is back and his leadership results could be evident by late 2015. Also AMD need their fab partners to deliver competitive process tech compared to Intel for high performance CPUs That is the bigger problem because the foundries are all bothered about low power process since mobile market drives demand for foundries. Intel dominates the server market because of excellent microarchitecture and also world class manufacturing process. AMD has to get Globalfoundries to work with IBM on FDSOI for high performance CPUs on future nodes.

I imagine he's there to work on the ARM stuff. It sure looks like Excavator/Big Core is dead.

 

[StrangerGuy]

AMD's marketshare is so low in servers that they can make a lot of gains even by addressing niches such as HPC, Berlin APU + GPGPU, and dense clusters, Seamicro acquisition + Cat family x86/64-bit ARM SoC. If those gain them just ~5-6% more over all server share then they've more than doubled their market presence.

AMD just does not have the management nor resources to compete head to head with Intel in the "traditional server market" and probably not in any market that doesn't value their GPU advantage or their willingness to offer lower priced product. Intel is in a different weight class than the rest of the competition due to maintaining the lead in fabrication. This won't change until Intel hits a fab wall that they can't spend their way past. There are already some signs that Intel may be slowing down in that regard due to market pressures (less Capex to power through on the fab side) and the increased technical difficulties of continuing to shrink transistors: http://www.reuters.com/article/2013/07/17/intel-results-idUSL1N0FN20G20130717

Exactly, they need avoid fighting pointless battles where Intel has dominated like raw CPU power and everything that they don't like "soft monopolizing" the gaming landscape with XB1/PS4.

 

[seitur]

AMD's biggest problem is their big core execution. the steamroller based kaveri is delayed to feb 2014. there are no steamroller based opterons for 2014 which is frankly weird.

Would reinforce an idea that Steamroller is directed at mobile market first and foremost and even though have better performance per 1 mhz, overally it does have significantly lower clocks.

Low-to-medium clocks, lower power consumption, better iGPU. - would kinda make good laptop chip, but nothing to boost about for other markets.

 

[Lonbjerg]

Why isn't the thread title : AMD Q2 2013 Market Share up 2.2% Q to Q but down 1.1% compared to 2012?

 

[Arachnotronic]

Why isn't the thread title : AMD Q2 2013 Market Share up 2.2% Q to Q but down 1.1% compared to 2012?

Quite simply, this is because people have a tendency to root for the underdog.

 

[Lonbjerg]

Quite simply, this is because people have a tendency to root for the underdog.

Reminds me of the Stockholm syndrome...not a very useless thing.

 

[wilds]

I'm rooting for AMD because we need a competitive environment!

 

[mavere]

Whenever Q/Q is reported instead of Y/Y, it's blatantly obvious that numbers are being spun a certain way.

 

[Vesku]

Isn't the main point of listing this Q to Q information that AMD actually gained some marketshare back? When was the last time AMD's marketshare was increasing?

 

[Lonbjerg]

Isn't the main point of listing this Q to Q information that AMD actually gained some marketshare back? When was the last time AMD's marketshare was increasing?

Last time was when they had comparable performance to Intel..that was YEARS ago.

 

[mrmt]

Last time was when they had comparable performance to Intel..that was YEARS ago.

AMD usually gains a bit of share once they launch new products. They gained some share in mobile when they launched Llanno and they gained share when they launched Brazos. IIRC the only products that didn't get them share were Bulldozer, Trinity and Vishera.

That said, a quarter doesn't mean anything, especially a quarter you launch a high volume product like Jaguar. That just tell us that OEMs are with high expectations for the product. Whether these expectations can become sales, that's another story.

Q2 had all elements for a smooth run for AMD, Q3 and Q4 are when things happen. Q3 results will show Haswell impact in the supply chain. Q4 will show Bay Trail impact.

 

[svenge]

Last time was when they had comparable performance to Intel..that was YEARS ago.

AMD's performance hasn't been comparable since July 27, 2006 (i.e. the launch date of Conroe-based "Core 2 Duo" chips). Seven years and counting...

 

[Lonbjerg]

AMD usually gains a bit of share once they launch new products. They gained some share in mobile when they launched Llanno and they gained share when they launched Brazos. IIRC the only products that didn't get them share were Bulldozer, Trinity and Vishera.

That said, a quarter doesn't mean anything, especially a quarter you launch a high volume product like Jaguar. That just tell us that OEMs are with high expectations for the product. Whether these expectations can become sales, that's another story.

Q2 had all elements for a smooth run for AMD, Q3 and Q4 are when things happen. Q3 results will show Haswell impact in the supply chain. Q4 will show Bay Trail impact.

My hunch is that they will lose marketshare in 2013 compared to 2012.
And the same for 2014 vs 2013.

 

[SiliconWars]

Q2 had all elements for a smooth run for AMD, Q3 and Q4 are when things happen. Q3 results will show Haswell impact in the supply chain. Q4 will show Bay Trail impact.

Looking through the recent threads here...

...can't find Kabini in laptops.
...can't find Temash in tablets.

Yet now it was a "smooth run" in Q2? Haswell launched one week after Kabini. Why is Q3 going to show a Haswell impact, did the extra week really make that much difference?

How will Bay Trail be showing an impact in Q4, when you can't even find Temash in a tablet? Is Bay Trail now supposed to harm AMD's non-existent tablet share? None of your points make sense.

 

[SiliconWars]

Whenever Q/Q is reported instead of Y/Y, it's blatantly obvious that numbers are being spun a certain way.

Tagging this for Q3 results. You'll understand why at that time.

 

[Lonbjerg]

Tagging this for Q3 results. You'll understand why at that time.

Bookmarked for later lulz.