AMD talked to NVidia about aquisition before getting ATI

[Topweasel]

Can one make the claim that AMD's APU strategy has been more effective than nVidia's mobile strategy so far? Considering AMD may have had limited resources, they spent them on what made the most sense for AMD at that time. AMD can still enter the mobile market and try to innovate SOC wins.

Nvidia's mobile plan was to start from the very bottom and work its way up. AMD and Intel have taken the exact opposite approach. Take what they have and pare it down to fit a Niche. Eventually Nvidia's ARM approach and their Atom/E-series will equal out. But the current market for AMD/Intel has almost disappeared because of pressure from companies like Nvidia and their ARM approach.

Now looking at AMD and their APU in the true PC market. It's a smart move. Going forward specially with the development of OpenCL and GCN, if developers take use of it can become a powerhouse chip. But we are 2 years off from a GCN based APU, and god knows when we will see enough OpenCL software to basically buy a video card as a CPU. But in the worst case the APU market can stay an effective slimtop and portable full power CPU/GPU. It will be what keeps AMD rolling.

 

[formulav8]

Well technically they aren't called anything been dead for half a year or so, and dieing for almost a full year before that.

Wow your right. I knew they were having a rough time but didn't know they actually croaked. :\

 

[railven]

Athlon 64 CPUs IIRC were not cheaper than Intel CPUs, and neither were the Athlon 64 X2 CPUs when they first came out...maybe Athlon and Athlon XP CPUs were.

Here are some X2 prices:
http://www.anandtech.com/show/1676

"Now armed with final silicon, our stance on AMD's Athlon 64 X2 doesn't change at all - AMD clearly has the faster overall dual core desktop solution, but at a price that will be out of reach for most users."

I never said AMD didn't have expensive products, they were still mainly recommended over Intel because of their price : perf in the lower brackets.

EDIT:
http://www.anandtech.com/show/1665/4

AMD's answer to Intel's aggressive pricing is two-fold. Eventually, all of AMD's CPUs will be dual core, and thus, prices will be driven back down to single core levels. But for now, AMD feels confident enough that their single core CPUs are fast enough to compete with Intel's low clocked Pentium Ds. We put that exact thinking to the test in Part II of our Intel dual core preview and concluded that it really depends on what type of a user you are. If you tend to multitask a lot or run a lot of multithreaded applications, then a slower Intel dual core is what you need; otherwise, a faster single core AMD is your best bet.

AMD single cores were relatively cheaper and still relatively better in price : perf for most users.

6 months later Core Duo launched and AMD lost the top. They still held the bottom, and today they're starting to lose even the bottom.

Regardless of that performance crown period, people will still remember AMD-CPU as the budget brand.

 

[blackened23]

Nvidia's mobile plan was to start from the very bottom and work its way up. AMD and Intel have taken the exact opposite approach. Take what they have and pare it down to fit a Niche. Eventually Nvidia's ARM approach and their Atom/E-series will equal out. But the current market for AMD/Intel has almost disappeared because of pressure from companies like Nvidia and their ARM approach.

The margins for ARM mobile products are ridiculously thin and nvidia has a worse product than Quallcom / TI. Nobody cares if a mobile product has gimmicky acronyms or whether it is quad core, they care about user experience. Aside from that, NV has locked themselves out of the x86 market which is still huge - AMD sold 30 million APUs last year. Apple products and ultrabooks are all x86 and that is still a huge market with comfortable profit margins.

I have mixed feelings on nvidia pursuing this market. Its a tough market to compete in with thin margins, and their product doesn't perform as well as it should. I'm sure nvidia can rectify this (performance) in future products -- right now tegra 3 doesn't fare well against iPad2 in graphics performance (performs much worse). ARM is probably their only choice, though -- NV is locked out from the x86 market.

 

[BallaTheFeared]

I wasn't aware that Tergra 3 was that bad, I had heard up to 3x the graphics performance of Tegra 2 but then I don't have a need for those types of products so I haven't been following them.

All I know is Tegra 2 sold decently and accounted for $360 million out of $4 billion in sales last year for Nvidia. I'm sure they'd like to grow that market share, but I'm not seeing the need for the doom and gloom present around here.

 

[SirPauly]

The margins for ARM mobile products are ridiculously thin and nvidia has a worse product than Quallcom / TI. Nobody cares if a mobile product has gimmicky acronyms or whether it is quad core, they care about user experience. Aside from that, NV has locked themselves out of the x86 market which is still huge - AMD sold 30 million APUs last year. Apple products and ultrabooks are all x86 and that is still a huge market with comfortable profit margins.

I have mixed feelings on nvidia pursuing this market. Its a tough market to compete in with thin margins, and their product doesn't perform as well as it should. I'm sure nvidia can rectify this (performance) in future products -- right now tegra 3 doesn't fare well against iPad2 in graphics performance (performs much worse). ARM is probably their only choice, though -- NV is locked out from the x86 market.

Can you offer what these margins are?

 

[notty22]

Nvidia market share is low with Tegra , but it is there participation , in part, why Microsoft is going forward to offer Windows for ARM.
NVIDIA Ushers in World's First Tegra 3 Super Phones, Delivering Outstand

It's part of there business that has helped make up for losing chipset revenue. Evolve,adapt or pay the price.

edit:
And the Oscar Goes to Hugo, Rango…and NVIDIA Quadro

 

[BallaTheFeared]

And the Oscar Goes to Hugo, Rango…and NVIDIA Quadro

"Winning"

 

[exar333]

You want to know what the biggest difference between ATi being part of AMD versus JHH running AMD? Take a look at this chart again-

http://www.techpowerup.com/img/12-02-25/jpr_mdgpu_2011_01.jpg

See the Qualcomm numbers? AMD sold them that technology for $65 Million. One of the most absurdly stupid business choices ever. nVidia has ~10% of that market share and is posting quarterly profits for that division higher then $65 Million. This is why the comparison to JHH versus whatever helmet head AMD has crashing(can't really say running) the business into the ground is laughable.



The SoC market isn't anything you seem to understand. Nothing is out that can compete against Tegra3 at the moment. Some of the chips that we have seen that are supposed to handily beat Tegra3 aren't scheduled to ship in a device until 2013- in some cases several quarters after Tegra4 is out. Until nV gets LTE integrated with the Tegra line they are going to have a rough go against Qualcomm. As of this moment however, nothing is shipping that can top Tegra3(Samsung has even pushed back the announcement of their product that is supposed to be better- not the ship date, just the announcement of products based on Exynos 5250).

BTW- The 3% doesn't include tablets which is by far Tegra's strongest market(well, tablets and automobiles, but automobile integration isn't a hot topic for fanboys, just people who are interested in the business end of making money).



Intel has been trying to do that for years. They find a way to fail.


They have done that already, it's called the Snapdragon and it is fairly dominant in the market. The morons who helm AMD just decided to sell it for $65 Million. One of the dumbest business choices in history.



Qualcomm and Samsung are the big boys in the SoC market. Apple doesn't make a SoC as of yet. You can read whatever charts you like, the A6 is the first SoC that Apple may be having made on their own(TSMC fabbing it for them). Samsung products all of the A5s.



nVidia? No, certainly not. nVidia makes money hand over fist in its' ventures, AMD finds a way to lose money no matter how great the product is. JHH certainly would have helped that out enormously, but I don't think it would have been enough to equal the companies out on a straight profitability metric.

A lot of signs appear to show that changing. Time will tell. Intel is much more likely than AMD to succeed in the mobile arena. AMD's chance came and went already on this front.

 

[at80eighty]

get to walk by paying back a mere percentage of your ill-gotten goods.

thats just wrong on more than one level. yikes

 

[Dribble]

Can one make the claim that AMD's APU strategy has been more effective than nVidia's mobile strategy so far? Considering AMD may have had limited resources, they spent them on what made the most sense for AMD at that time. AMD can still enter the mobile market and try to innovate SOC wins.

APU is case in point. It current state is:
a) it's still just a cpu with on-chip graphics, the apu part is still marketing. It should be an APU but AMD hasn't got there yet and isn't going to for years.

b) to make an APU you need not only hardware but software. All AMD has is opencl and that is being developed far to slowly because no one is really pushing it. It can't take off without the software, AMD needed to invest far more in this - they can't just write some drivers and assume the rest will magically appear.

c) forget the APU and think of just cpu with graphics which is what we have now. That arrived years to late. AMD bought Ati and had this as a target for a long time. They should have beaten intel to market and stolen lots of sales, as it was Intel beat them to market by at least a year! This meant by the time llano appeared it was just about enough to hold off intel, it wasn't going to take over.

It sums up AMD for the last few years - there's a great product in their somewhere but it never arrives on time, and is always cut back or incomplete and unable to match expectations. There is far to much pointing at marketing slides and talking about what uber product might be about to appear, and far to little buying that uber product today.

 

[SirPauly]

Imho,

There is no doubt there is more innovation and work to do with the APU from the hardware and software side but the key to me is how much revenue, profits and margins have AMD enjoyed with their APU strategy so far?

One may make the claim the APU turned their compute sector into a profitable sector and there is still growth promise and innovation potential.

 

[pw38]

Excuse me, I have no problem with the comment in itself. Its where the comment is being made. Thankyou.

I didn't say you had a problem with the comment only that he had a point in calling you out on what you said. You said it first and he pointed it out, nothing more. Anyway no worries. Don't mean to keep off topic.

 

[bryanW1995]

That is pretty much commonly known. Jen wanted to be ceo no matter what. As much as I dislike the guy he would have been better than Ruinz

AMEN!!

Unfortunately, back in those days AMD had around twice the market cap of nvidia, so I can see the reluctance of the AMD brass to let the acquiree become CEO. On the other hand, I can't think of anybody I'd rank higher than jhh on my "tech CEO" draft board. And, of course, AMD got smoked on that ATI acquisition.

 

[bryanW1995]

I have taught business ethics to business students (and students in many other majors). Many of the students express open contempt for the content.

Generally speaking, morals/ethics are taught in the home, not the classroom. An individual's intellect and motivation sadly have very little to do with that individual's moral compass.

 

[Topweasel]

The margins for ARM mobile products are ridiculously thin and nvidia has a worse product than Quallcom / TI. Nobody cares if a mobile product has gimmicky acronyms or whether it is quad core, they care about user experience. Aside from that, NV has locked themselves out of the x86 market which is still huge - AMD sold 30 million APUs last year. Apple products and ultrabooks are all x86 and that is still a huge market with comfortable profit margins.

I have mixed feelings on nvidia pursuing this market. Its a tough market to compete in with thin margins, and their product doesn't perform as well as it should. I'm sure nvidia can rectify this (performance) in future products -- right now tegra 3 doesn't fare well against iPad2 in graphics performance (performs much worse). ARM is probably their only choice, though -- NV is locked out from the x86 market.

All of this is highly debatable. The Tegra 2 and Tegra 3 beat competitors out by almost 6 years. The original Tegra didn't sell well, but again you could get a 600Mhz Tegra 1 Zune almost a year before the TI in the OG Droid.

Now Nvidia needs to work on the graphics of their chips. But as it is they have sold millions of chips to Motorola, Acer, Asus, and even a few to Samsung. By basically beating everyone to the punch. The T3 all things considered as a 40nm 5 core CPU is actually pretty spectacular. Sure as the Krait and other 28nm chips start gaining speed the T3 will start to age, but they have basically taken the ATI approach as of late by being early each time. I would have to say when not having the ties to phone providers (Samsung, HTC/qualcom, Motorola/TI) they have actually done a fantastic job.

 

[notty22]

http://www.youtube.com/watch?v=lUFv4C4fVzE&feature=player_embedded

Go to the 54.00 minute mark, and you see Windows 8 running on 4 ARM platforms, the one showcased live is the Tegra 3. Interesting listen.

 

[bononos]

Not to perpetuate the off-topic nature, but it begged an observation. The tolerance and absolution for the crooks of the business world can ultimately be laid at the feet of the voters in countries where the government creates a climate that allows corruption.

Speak up and take notice with your vote, if you account for the majority, things will change. Find you're not in the majority and things stay the same, immigrate

Mere voting is only going to have a marginal effect at best on crooked behaviour. If the system is set up to fail as with the neoliberal thinking that has pervaded our common sense and intellectual space for the last few decades then elections are going to be only a method for crooked elites to compete among themselves.