AMD Piledriver rumours ... and expert conjecture

[Reynod]

We have had several requests for a sticky on AMD's yet to be released Piledriver architecture ... so here it is.

I want to make a few things clear though.

Post a question relevant to the topic, or information about the topic, or it will be deleted.

Post any negative personal comments about another user ... and they will be deleted.

Post flame baiting comments about the blue, red and green team and they will be deleted.

Enjoy ...

 

[jimmysmitty]

Never say "True That" again please. This is not Ebonics class. I think. There are no rumors about a delay of Trinity. They are already sampling them. If you look around it's mainly Marketing people, NOT engineers. That's like MS firing all of the developers.

I'm pretty sure someone said they were sampling BD in April/May and would release in June.

Sampling does not always mean its ready. They could run into any number of issues and delay it, just as they did with BD.

 

[Reynod]

signorechubbs, 384-bit and jdwii I give the three of you fair warning ... publically ... and for the benefit of others ... I viewed the deleted threads above this post and should you post one more personal slight against another user I will give you a 3 day ban and remove all your subsequent posts from this thread.

None of the moderators will tolerate any more bullying nonsense either.

Mal, if you want a sticky in the graphics area post a request down there with the CR concerned ... don't derail this topic please.

 

[gamerk316]

If it can be shown that using a kludge like you advocate has absolutely no negative impact on future performance in a variety of situations then it should be implemented. If that can not be shown beyond a shadow of a doubt then that solution should not be implemented and we just have to wait until a kernel patch is ready.

The *kludge* in this case is simple: Use the cores with the 20% performance penalty LAST. Simple fix to the scheduling problem.

If AMD wants to broadcast an 8 core CPU, thats their problem. Just don't whine and complain when the OS doesn't use it properly.

 

[Reynod]

I said nothing against any user at all reynod...

Does "True that" or "Amen" come to mind?

I have all the deleted threads and the edited comments ... as do the other mods / CR's ... and the CR in particular targeted.

Enough said ... moving on.

 

[aford10]

As you are a responsible moderator i suggest you take a look at this thread: http://www.tomshardware.co.uk/forum/317613-10-battlefield
I put in a request for a moderator ages ago.

Taken care of.

 

[aford10]

can I get some help.?
much appreciated.

Not sure I can offer the kind of help you need. 😛

But if there's something that you want looked at, use the report link, or send me a PM.

 

[keithlm]

The *kludge* in this case is simple: Use the cores with the 20% performance penalty LAST. Simple fix to the scheduling problem.

If AMD wants to broadcast an 8 core CPU, thats their problem. Just don't whine and complain when the OS doesn't use it properly.

I see you have chosen to ignore all regression testing issues mentioned above. It would be nice if everything was as simple as you are pretending because then we'd see a lot less of people complaining about the issue. (This would include you.)

It is still slightly possible that you are correct and they could release a quick bios patch. Not likely; but possible.

 

[fazers_on_stun]

Never say "True That" again please. This is not Ebonics class. I think.

Even though I am Caucasian, I think your remark is uncalled for and racially insensitive. Please edit it out of your post, thanks.

There are no rumors about a delay of Trinity. They are already sampling them. If you look around it's mainly Marketing people, NOT engineers. That's like MS firing all of the developers.

From http://www.brightsideofnews.com/news/2011/11/3/exclusive-letter-from-amd-ceo-rory-read-to-the-company.aspx

The cuts however, did not touch just the marketing and sales staff. Key Fusion engineers were let go as well.

Isn't Trinity = PD + Fusion? If so, then it looks like at least part of it will be affected..

As the third paragraph states, AMD will announce its new course, codenamed "Project WIN"on November 9th, 2011.

I guess we'll find out in 5 days what AMD's 'new course' will be. I would guess even less emphasis on DT and more on mobile, or who knows - maybe they'll go up against ARM on cellphones 😛.

 

[keithlm]

Actually I'm the one who posted "True that"

OFF TOPIC: If you look that term up in various slang dictionaries you find that the term is now used primarily by juvenile males of caucasian descent.

 

[gamerk316]

I see you have chosen to ignore all regression testing issues mentioned above. It would be nice if everything was as simple as you are pretending because then we'd see a lot less of people complaining about the issue. (This would include you.)

It is still slightly possible that you are correct and they could release a quick bios patch. Not likely; but possible.

As someone who's done his share of hardware regressions in the past, I view the schedulng fix I described as nothing more then a core loading function. Use the cores that scale to 100% before the ones that scale only 80%. Run through a few regressions that use different workloads to confirm no problems, and ship the fix.

Instead, you'd rather MS kludge their scheduling logic for a processor architecture that is basically lieing to the OS about its capabilities, rather then AMD actually do something about the fact its processor is lieing about how effective its cores are.

Call a physical core a physical core, and call a logical core a logical core. But if you call a logical core like a physical core, don't complain when the OS happily uses it.

 

[biggtex]

I'm loving my bulldozer rig I just scored 8100.21 on the passmark I will have it post here in the next few days I still have to do a few things to be able to get a better score

 

[biggtex2011]

yea that is my uncles system we are on my bulldozer right not and i'm getting mine up there in the next few months

 

[JAYDEEJOHN]

If there were to be a delay, which you cant have unless its way late, the cause wouldnt be the layoffs, as most are coming to the newest groups, not more seasoned employees, tho not exclusive to time of hire.
The delay would be because, the fixes and the process isnt ready, tho Im sure the process end will be done.
They wont make the same mistake twice, PD will be ready when it is ready, delay or no delay

 

[dogman_1234]

If AMD delays...BD won't be their Only problem...

They cannot afford to delay PD or Trinity at all.



I would have to say the layoffs are not a good sign. I read somewhere,( so do not quote me on this,) that they will only save 28 mill USD on the 10% layoff.

I will try to see where that was.

 

[BadTrip]

FX8150 vs. 2500K with 2-way/3-way SLI

http://www.hardocp.com/article/2011/11/03/amd_fx8150_multigpu_gameplay_performance_review/1

 

[JAYDEEJOHN]

This year, the layoffs wont amount to much, but next year and counting, Ive heard 118 million to 200 million per year

 

[biggtex2011]

FX8150 vs. 2500K with 2-way/3-way SLI

http://www.hardocp.com/article/2011/11/03/amd_fx8150_multigpu_gameplay_performance_review/1

Yes but there had been better bios put out during and for cross there are prototypes for cross fire even more so if you work for amd like my uncle does.

 

[keithlm]

Instead, you'd rather MS kludge their scheduling logic for a processor architecture that is basically lieing to the OS about its capabilities,

You are incorrect in your assumption of my opinion. Personally I'd rather just see the correct patch be released. I don't want to see any band-aids and half measures that might work but can't just be just thrown out as easy as you believe.

BTW: The word is spelled "lying" not "lieing". Which is a word that is not appropriate for this conversation unless you are merely trolling. (Which you shouldn't be doing.)

 

[BadTrip]

Yes but there had been better bios put out during and for cross there are prototypes for cross fire even more so if you work for amd like my uncle does.

Coherency please. You are from Texas so English should not be a problem.

 

[jimmysmitty]

As you are a responsible moderator i suggest you take a look at this thread: http://www.tomshardware.co.uk/forum/317613-10-battlefield
I put in a request for a moderator ages ago.

Hah. Wonder if that guy tried installing VF3 on his machine....

Actually I'm the one who posted "True that" in response to Archibael's comment about the AMD layoffs being bad in this economy (and possibly having ramifications for Trinity delays).

However, as long as we're off-topic, let me comment that I find Baron's "Ebonics" joke or whatever to be racially insensitive, if not outright offensive. I trust he operates under the same TOS as the rest of us do?

I actually agree. The others didn't really do anything except assert their freedom of speach but his comment was a tad off.

FX8150 vs. 2500K with 2-way/3-way SLI

http://www.hardocp.com/article/2011/11/03/amd_fx8150_multigpu_gameplay_performance_review/1

Wow. Interesting article.

Yes but there had been better bios put out during and for cross there are prototypes for cross fire even more so if you work for amd like my uncle does.

I doubt a BIOS will improve the performance there as its GPU limited. But still its possible as thats a pretty big performance gap for only 200MHz clock speed difference.

I guess we will see in the future.

 

[Mousemonkey]

I was thinking of a overclock section for nvidia/AMD gpu's too as people often come here asking about this.

What? Like this one that already exists?

http://www.tomshardware.com/forum/forum-29-181.html

 

[jimmysmitty]

If any mod knows why the message I restored was deleted, please let me know. From what I can see it was still within lines of this thread, it had a link to info of Trinity, Trinity is a PD part so its on topic, and as well a theory on fazers part that Trinity might get delayed.

 

[Mousemonkey]

If any mod knows why the message I restored was deleted, please let me know. From what I can see it was still within lines of this thread, it had a link to info of Trinity, Trinity is a PD part so its on topic, and as well a theory on fazers part that Trinity might get delayed.

Not guilty, so I dunno.

 

[JAYDEEJOHN]

Guessing here, it was too much about gpus?
Wasnt me

 

[fazers_on_stun]

^ Well I have a pretty good guess as to who deleted it. It may be time to start a mass petition to jpishgar.

Back on topic: http://www.xbitlabs.com/news/other/display/20111103135021_Advanced_Micro_Devices_Set_to_Unveil_New_Strategy_Next_Week.html

AMD wants to put more efforts into consumer-oriented product designs, another person, who also asked not to be identified, with knowledge of the plans revealed. In particular, the code-named Bobcat micro-architecture, which powers the company’s popular Brazos platform, will play a much more important role in the company’s future that it was previously planned. In approximately one year the company shipped 20 million Brazos (Zacate, Ontario, etc.) chipsets for netbooks and low-power notebooks, a large number. Going forward, the firm will attempt to introduce flavours of Brazos aimed at tablets, consumer products and other devices.

What is unclear is what AMD plans to do with its high-performance and server offerings. While it is hard to expect the company to completely get rid of them, it is believed that the company will change its corporate targets and rebalance investments.

This is in-line with AMD's cost-cutting of up to 1500 employees, as the low-end parts have lower margins and even with a lot of volume, they simply don't generate as much cash. So, you could say Bulldozer has a cache problem and Bobcat has a cash problem 😛.