• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD Q415 results

Page 9 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
You should check how much they have sold off over the time. And how many times they just created new shares.

Mubadala cares nothing for AMD. For them AMD is a milking cow (WSA).

Your concern and care about AMD is noticeable in the internet, since 2012 every hour, every day, like a paying job.
 
How would the ATI business look like today, if it wouldn't have been bought?

Much better. I would say a company with smartphone GPUs and a good competitive setting against NVidia with a solid R&D. The main problem for ATI with the AMD purchase is, that it has always been second class in a company that focus on CPUs. Specially server oriented CPUs. ATIs job in AMD was to become an IGP add on.
 
Lisa Su said on the call that Zen-based server chips will be in the performance class of ~80% of the server market; won't be fast enough for the top 20%.

Given the fact that Zen would be using Multi Chip Module to achieve the highest core counts (32C/64T) while Intel would be using single die to hit the highest core counts on Skylake Xeon EP(28C,56T) AMD is not going to be selling any 8 way systems at all. Rather their 4 way systems would actually be 8 way.

http://www.qdpma.com/systemarchitecture/SystemArchitecture_Opteron.html

Also AMD's RAS features are likely to be much more limited than Intel's who has been improving them for a long time from Nehalem. So AMD is not going to sell 8 way systems and their appeal in 4 way servers will be limited as Intel will have the better feature set with the Xeon EX. This leaves AMD Zen competing with only Xeon EP. I think thats the reason behind AMD being able to address only 80% of the server market.
 
They will release A Zen CPU on December 31st, 2016 🙂
Yes, or at least late 2016. Then larger volumes in 2017Q1. I wouldn't be surprised if that happens. Seems to be common practice often, for nVidia, Intel, AMD, ... But it doesn't matter much anyway if it's late 2016 or early 2017 I think.
 
"only" 80%. Seems like quite a big share of the market to me anyway... :hmm:

Yes, Fjodor, they're going to capture all of this share and AMD will be tops in the server market. You will finally get your wish and Intel will crumble, leaving the world with only the finest 8 core Zen processors for $200.
 
Yes, Fjodor, they're going to capture all of this share and AMD will be tops in the server market. You will finally get your wish and Intel will crumble, leaving the world with only the finest 8 core Zen processors for $200.

😉 That was not my point, I thought that was obvious. The point is that they have a shot at a part of 80% of the server market. How bit part of that they'll get we don't know. But at least Zen is a candidate for a large portion of the server market, which is what's interesting.
 
😉 That was not my point, I thought that was obvious. The point is that they have a shot at a part of 80% of the server market. How bit part of that they'll get we don't know. But at least Zen is a candidate for a large portion of the server market, which is what's interesting.

It's interesting that Forrest Norrod was talking at Analyst Day about how unlike Intel, AMD's low share allows it to target specialized areas of the server market instead of having to build a big portfolio of products to serve just about everything.

AMD management, seemingly desperate to pump their stock in light of a terrifying financial reality, are now hyping up how they will have SKUs good for 80% of the market?

What a joke.
 
It's interesting that Forrest Norrod was talking at Analyst Day about how unlike Intel, AMD's low share allows it to target specialized areas of the server market instead of having to build a big portfolio of products to serve just about everything.

AMD management, seemingly desperate to pump their stock in light of a terrifying financial reality, are now hyping up how they will have SKUs good for 80% of the market?

What a joke.
Ok, so what is it you have to object to the 80% claim? Are you saying it is not valid, and if so for what technological reasons?

Do you know the performance, metrics and pricing of Zen to be able to make such a judgment for certain?
 
Much better. I would say a company with smartphone GPUs and a good competitive setting against NVidia with a solid R&D. The main problem for ATI with the AMD purchase is, that it has always been second class in a company that focus on CPUs. Specially server oriented CPUs. ATIs job in AMD was to become an IGP add on.
Looking at it with what we now know about how the markets changed, this sounds plausible. The mobile GPU business likely wouldn't have been sold to Qualcomm, but further licensed to them. For consoles the combination of GPUs and processors from different IP vendors has been done multiple times, so no further problem. The compute part might have developed differently. BTW, if Larrabee would have been a success, even Intel might have focused more on such parallel compute capabilities, driven the environment and standards. And maybe, they wouldn't pack these huge SIMD units into their CPUs now. 😉 But they had to, as long as heterogeneous computing didn't thrive the markets.
 
Also AMD's RAS features are likely to be much more limited than Intel's who has been improving them for a long time from Nehalem. So AMD is not going to sell 8 way systems and their appeal in 4 way servers will be limited as Intel will have the better feature set with the Xeon EX. This leaves AMD Zen competing with only Xeon EP. I think thats the reason behind AMD being able to address only 80% of the server market.
They put a lot of work into RAS. Wonder, how it turns out to be.
 
It's interesting that Forrest Norrod was talking at Analyst Day about how unlike Intel, AMD's low share allows it to target specialized areas of the server market instead of having to build a big portfolio of products to serve just about everything.

AMD management, seemingly desperate to pump their stock in light of a terrifying financial reality, are now hyping up how they will have SKUs good for 80% of the market?

What a joke.

Selling ~30% compared to Intel into the 80% of the x86 server market AMD claims Zen can compete in would roughly double AMD's revenue.
 
Much better. I would say a company with smartphone GPUs and a good competitive setting against NVidia with a solid R&D. The main problem for ATI with the AMD purchase is, that it has always been second class in a company that focus on CPUs. Specially server oriented CPUs. ATIs job in AMD was to become an IGP add on.

This 100%. The writing was on the wall when ATI's CEO walked away.

AMD, ATI, consumers would all have been better off with a licensing deal rather than an acquisition.
 
This 100%. The writing was on the wall when ATI's CEO walked away.

AMD, ATI, consumers would all have been better off with a licensing deal rather than an acquisition.

Yep. Sad development. But the sky was the limit. And they still think it is.
 
This 100%. The writing was on the wall when ATI's CEO walked away.

AMD, ATI, consumers would all have been better off with a licensing deal rather than an acquisition.
ATI was about to die... if AMD didn't got them, Intel would do.
And you know what will happen if that event happen.
 
ATI was about to die... if AMD didn't got them, Intel would do.
And you know what will happen if that event happen.

Ruiz paid too much for ATI because he wanted to stay CEO and also get a nice stock vesting from a merger. JHH wouldn't agree to an AMD-Nvidia merger unless he retained CEO seat. No matter how many books Ruiz writes, this was clearly what happened. AMD was then left with little capital to keep fab improvements going nor fund the increase in R&D needed to integrate the two companies technology into SoCs and APUs in a reasonable timeframe.
 
AMD management, seemingly desperate to pump their stock in light of a terrifying financial reality, are now hyping up how they will have SKUs good for 80% of the market?

What a joke.

Do you feel like a joke yourself?

Because they didnt say anything like you write it.

I mean if you have low self esteem dont blame amd management.


insulting other members is not allowed
Markfw900
 
Last edited by a moderator:
Ruiz paid too much for ATI because he wanted to stay CEO and also get a nice stock vesting from a merger. JHH wouldn't agree to an AMD-Nvidia merger unless he retained CEO seat. No matter how many books Ruiz writes, this was clearly what happened. AMD was then left with little capital to keep fab improvements going nor fund the increase in R&D needed to integrate the two companies technology into SoCs and APUs in a reasonable timeframe.
Amd nvidia merger would have been far better for the industry. With jhh as ceo. He and nvidia got all that amd didnt have.
Market driven. Innovative from a consumer perspective. Marketing, pr technical marketing, all in spades. And they have always had since tnt. Besides jhh is energetic and knows his tech. Passion for tech and what you could do with it. Like steve jobs.
 
This 100%. The writing was on the wall when ATI's CEO walked away.

AMD, ATI, consumers would all have been better off with a licensing deal rather than an acquisition.

They also could have bought ImgTec for a fraction of the price and built it up from there. Or some other small company like what ARM Holdings did (ARM didn't go broke buying graphics IP and now they have arguably some of the best mobile GPU IP out there).
 
There was no limit on how Skylake performed.... until release. 🙂

Most important architecture in a decade and what not. 🙄

Skylake is the best there is. The product you champion is...unknown at best 🙂

The quote you refer to:

Intel believes that Skylake, the latest in its Core series, is the most significant processor it has released in the last decade. It’s based on a new circuit design and promises significant increases in performance, battery life and power efficiency.

The bar keeps moving, you better hope your Zen product can follow. Speedshift for example? I dont think so.
 
Last edited:
Back
Top