And what product do you think it's paying the bills today, the big core products that have been losing money and share every segment they are in, or the cat cores, which offer good value proposition compared to Intel's offers and have been opening *new* markets for AMD? AMD management would be very dumb if they were not to pursue something the cat core line, even if it's a little beefier than current cat cores.
Are we speculating or do you have numbers to share with us ??
https://owl.english.purdue.edu/owl/resource/589/2/There are, of course, certain things that do not need documentation or credit, including:
- When you are using "common knowledge," things like folklore, common sense observations, myths, urban legends, and historical events (but not historical documents)
- When you are using generally-accepted facts, e.g., pollution is bad for the environment, including facts that are accepted within particular discourse communities, e.g., in the field of composition studies, "writing is a process" is a generally-accepted fact.
What's the point of AMD achieving what are you describing with FX and imploding its balance sheet in the process? Because this is the most important achievement of the FX line, not developing a product that can match Intel's mainstream on raw performance (despite the asinine power consumption).
And what product do you think it's paying the bills today, the big core products that have been losing money and share every segment they are in, or the cat cores, which offer good value proposition compared to Intel's offers and have been opening *new* markets for AMD? AMD management would be very dumb if they were not to pursue something the cat core line, even if it's a little beefier than current cat cores.
You do understand you are quoting from a guide on how to avoid plagiarism, don't you?https://owl.english.purdue.edu/owl/resource/589/2/
You don't need to question things that you know are true, just because you don't like the guy and want to crap on his argument.
I don't think they would have to go bankrupt producing it. They have many good 'off the shelf' technologies that they can use and/or improve on. I think their focus should be on mobile, maybe a beefier cat core as you say, and then scale up from there for the desktop parts. But that's just me playing armchair CEO.
Taking the cat core route is not just going for a smaller core, but going for a simpler design, with heavy use of synthesis tools and focus on low cost and TTM, not on extreme raw performance. So if they take that route, they are automatically out of the performance race.
https://owl.english.purdue.edu/owl/resource/589/2/
You don't need to question things that you know are true, just because you don't like the guy and want to crap on his argument.
Seriously now, what is making more money for AMD? What has been a greater success? Their big cores, or their little cores?
The major question is, what are they targeting for with the new mArch.
I remember when Dothan was pretty competitive to AMD's mobile. Intel built on that using some of the good parts of the P4 as well as some new engineering altogether. I don't know that AMD could do that with the cat cores as they are far behind something like an i7. But, I could see them using lessons learned from both as well as some of the existing 'good' tech from both, and some fresh engineering to make a more competitive desktop level part.
As a note, Steamroller has a higher degree of design automation (less custom macros) as compared to Bulldozer IIRC, while Excavator was said to push it further by adopting high density cell library used in Cat cores and GPUs. Perhaps the new core following this path would be out of the race in the gimme-the-crown-with-highest-clock segments, but it might still be able to compete better in those they didn't do well since Bulldozer was out - IF they still intend to compete in those with the new core.Taking the cat core route is not just going for a smaller core, but going for a simpler design, with heavy use of synthesis tools and focus on low cost and TTM, not on extreme raw performance. So if they take that route, they are automatically out of the performance race.
It was mostly a joke. However the point that I was making is that you don't have to provide sources for common sense.You do understand you are quoting from a guide on how to avoid plagiarism, don't you?
Keep in mind the second meaning of the word "documentation": the process of classifying and annotating texts, photographs.
But that's of course just a guess based on what info is known so far. It would be nice if AMD could be more precise in communicating their intentions with the new x86 uArch going forward. So I wonder when they are likely to publish more detailed info? 2014/2015/2016?
Well both AMD and Intel have been a lot more secretive lately, and don't give out much details far in advance. I wonder if that's because they don't want to make any plans public too early, so that the competitors can make use of that and adjust their strategy accordingly?
That's a generally preferred approach for non-NDA customers. Project timelines, expected cost, performance, etc... You have to balance performance with cost/schedule. If you know competition is going to blow you out of the water, I'm sure some costlier ideas will get taken off the back burner.
https://owl.english.purdue.edu/owl/resource/589/2/
You don't need to question things that you know are true, just because you don't like the guy and want to crap on his argument.
Seriously now, what is making more money for AMD? What has been a greater success? Their big cores, or their little cores?
Well both AMD and Intel have been a lot more secretive lately, and don't give out much details far in advance. I wonder if that's because they don't want to make any plans public too early, so that the competitors can make use of that and adjust their strategy accordingly?
I agree. But that was also true ~5+ years ago, but back then both AMD and Intel were more open with their plans. So what has changed since then?
Experience.
People have become older and wiser, and carry forward with them the lessons learned of the past.
Who wants to keep getting scooped?
FX8350 vs Core i7 3820 have almost the same die size, almost same MT performance and close the same power consumption, both at 32nm.
The only problem of bulldozer is they used the same die for Server and Desktop. Intel is using a different die for Desktop (Core i7 2600) and different die for Server (Core i7 3820). Intel can do that because they sell higher volume and they can spend more for different masks. AMD only use one mask because of the smaller volume. That affects performance, power usage and economics.
Die size alone means nothing, a 300mm2 die on 32nm is cheaper than 200mm2 or smaller size on 14nm today.
I'm not sure why you are comparing a fully enabled chip to one where fully half the cores are disabled and a good amount of cache. You are also looking at a quad channel memory controller. If you are going to go by technical requirements you must look at the same die.
(blah).
3820 has a separate die. Transistor count is 1.27B and die area is 294mm2 as opposed to 2.27B/435mm2 for sandy 6c which is really an 8 core Xeon . It's still bigger then a regular sandy 4C that is 995M/ 216mm2
I'm sorry, I thought you were just busting his balls.I dont have the data, thats why i made a question. Does he have data to show us or is he speculating ??
^^ This.
And both AMD and Intel agree that 4 cores are plenty for non server/workstation workloads.