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 ...

 

[gamerk316]

That was the whole point of their blind test, to both see for themselves and demonstrate to others that the actual "differences" between PC components today are so small, that you'd never notice them without running a benchmark.

But heres the thing: One CPU will go obsolete quicker then the other one. If both perform about the same, why get the one that will be obsolete sooner? Nevermind the other one costs more, runs hotter, and is going to be on a different socket in very short order.

But hey, if you don't do anything aside from light desktop and internet, even a lowly C2Q would still look competitive. How about we blind test that as well?

 

[truegenius]

lol c2q will be a overkill for internet

 

[gamerk316]

^^ I only upgraded from my QX9650 because both my mobo and GPU went at the same time. It was still more then sufficent for my needs.

 

[esrever]

couldn't they just right click on the computer icon and find out the system details? that alone should reveal most of the information one needs. sometimes game options show system information. i don't think keyboard usage/access was restricted.
that's the first thing an informed user would do on an unfamiliar pc.... i think.

I doubt they let people exit the game to check the hardware specs. anyways if people checked system details, it would be a blind test...

 

[truegenius]

^+1

 

[gamerk316]

And in other news:

http://www.dailytech.com/article.aspx?newsid=23861

AMD aired its earnings for (calendar) Q4 2011 late Tuesday and the result was a clear miss. Versus much of the rest of 2011 [1][2][3], where it posted profits, Q4 saw a $177M USD net loss when the results were adjusted to the general acceptable accounting practices (GAAP), the U.S. accounting standard.

Revenue stayed constant from Q3 2011 at $1.69B USD. This is a fairly substantial miss from the analyst consensus earnings target of $1.71B USD [source].

Aside from a slightly depressed gross margin (down 1 percent), the net loss comes largely due to a set of charges (losses). AMD took a $209M USD impairment charge on its investment in the GlobalFoundries chip fab, a $24M USD payment to GlobalFoundries charge, and a $98M USD general restructuring charge. AMD began laying off some employees in Nov. 2011.

AMD warns that the situation will get worse, with weakening demand expected for H1 2012. It's predicting$1.56B USD in Q1 2012 revenue. Previously, analyst targets had hovered around $1.7B USD, but they've since been adjusted down to a slightly more optimistic $1.59B USD.

The chipmaker saw graphics revenue dip 10 percent, with mobile GPU sales down. This dip may be compensated in H1 2011 by the official availability of the Radeon 7000 HD series (codenamed Southern Islands). In the good news department, AMD's "computing solutions" department surged 7 percent in sales, keeping revenue steady from Q3 2011. AMD says its server and chipset sales have improved.

In total, AMD made a net income (annual profit) of $491M USD in 2011, up modestly from 2010, a critical turnaround year for the firm.

Coincidentally, shares are up about 3% on the day. Go figure...

 

[palladin9479]

But heres the thing: One CPU will go obsolete quicker then the other one. If both perform about the same, why get the one that will be obsolete sooner? Nevermind the other one costs more, runs hotter, and is going to be on a different socket in very short order.

But hey, if you don't do anything aside from light desktop and internet, even a lowly C2Q would still look competitive. How about we blind test that as well?

Your assuming that the majority of the consumer market takes these questions into account prior to purchasing.

They don't, the look of the case sells more then the type of CPU used.

 

[truegenius]

http://news.softpedia.com/news/AMD-Moving-to-Tick-Tock-Micro-Architecture-Model-Some-Say-248736.shtml
not so interested

 

[gamerk316]

Your assuming that the majority of the consumer market takes these questions into account prior to purchasing.

They don't, the look of the case sells more then the type of CPU used.

And you know what? Thats fine. But thats no reason to recommend a FX over a SB. SB is clearly superior, and the P/P ratio is on its side. So lets stop the baseless "it has more cores" and "it will be better in the future" arguments to try and justify the purchase.

SB is clearly better, so we recommend SB.

 

[-Fran-]

do they have the resources to pull it off and be successful..?

They don't have to do it with the same time frames as Intel, but it is weird for them to take this approach TBH.

What is Mr. Read thinking, really.

Cheers!

 

[g4114rd0]

- "I have a hops feeling this will be a lot of fun in the next two years." 

This h@ppens Intel graphics integrated can takes a part of highest profit market.

 

[palladin9479]

And you know what? Thats fine. But thats no reason to recommend a FX over a SB. SB is clearly superior, and the P/P ratio is on its side. So lets stop the baseless "it has more cores" and "it will be better in the future" arguments to try and justify the purchase.

SB is clearly better, so we recommend SB.

Maybe you have me confused with someone else. At no point in time have I ever recommended BD. I've gone so far as to recommend people not buy it due to it's poor value. Nor have I made any positive "it must be better blah blah" statements. I'm not BM.

I've merely said that the vast majority of consumers would never recognize the difference between a BD and a SB system, all other components being equal. So using the CPU type as a selling point is mute, at least in mass market OEM box's. Now gaming and enthusiast machines are a different story, but they represent a minority of the PC purchase's made.

 

[esrever]

AMD needs to really step up their cpu game as the win7 hotfixes didn't help much at all.

I would have expected more than a 3% increase but Its looking like its pretty crappy.

 

[jimmysmitty]

do they have the resources to pull it off and be successful..?

I doubt it.Even being "FAB Lite" doesn't help enough to do it like Intel. Intels is based on a yearly schedule, I doubt AMD could keep up with Intel on that. Also considering Intel is puttin $5Billion into a 14nm FAB in Chandler (just an hour away from me) meaning $5Billion goes into jobs here in the US and then even more when the FAB opens.

And yes, the hotfixes were just nothing special. From what we saw with the Windows 8 preview, it shouldn't be suprising. While Windows 8 will be a bit better, it was nowhere near where the claims were and should not have been expected for Windows 7 either.

AMD needs a new efficient arch thats on a efficient process node. I have a feeling that wont be until at least 22nm for them. Could just be AMDs bad luck. 65nm was pretty poor for them, so is 32nm. Maybe its every other process node for AMD that will be decent.

 

[viridiancrystal]

http://news.softpedia.com/news/AMD-Moving-to-Tick-Tock-Micro-Architecture-Model-Some-Say-248736.shtml
not so interested

"All in all, the Trinity APUs, based on early Piledriver, will be superior to the first Bulldozer but not as advanced as the CPUs powered by the "full" Piledriver."

Trinity is going to be better than bulldozer? Anyone else catch that?

 

[gamerk316]

^^ And...? If BD is weaker then the PII, and Llano was powered by a PII based processor, and Trinity is more powerful then BD, then one could argue that Trinity is going to be another bust.

I'm very skeptical, simply because of the arch they based Trinity on.

 

[-Fran-]

^^ And...? If BD is weaker then the PII, and Llano was powered by a PII based processor, and Trinity is more powerful then BD, then one could argue that Trinity is going to be another bust.

I'm very skeptical, simply because of the arch they based Trinity on.

Well, if you think that Llano nor Trinity are meant to be "power computing" solutions, I see no harm in them having less muscle than the "pure enthusiasts" CPUs.

Making a comparison against something not meant to be compared to is a moot point.

I'm do worried about Trinity not being better overall than Llano, though. I couldn't care less if Trinity doesn't beat my PhII or the FX8150 CPU wise, but if it doesn't beat my Llano A8, then I'll be really mad.

Cheers!

 

[truegenius]

if it doesn't beat Llano A8 then it will be a proof of very very bad arch of bd (as pd is ebd i need atleast 10-15% improvement cpu wise)

 

[-Fran-]

But I'm not talking about CPU wise alone. The GPU component is important as hell also. But they work as a pair, so neither part has to slack to be "overall better".

Cheers!

 

[bawchicawawa]

^^ And...? If BD is weaker then the PII, and Llano was powered by a PII based processor, and Trinity is more powerful then BD, then one could argue that Trinity is going to be another bust.

I'm very skeptical, simply because of the arch they based Trinity on.

Simple, llano wasn't powered by a PII processor, it has a modified athlon chip in it, get your facts straight before passing judgment...

 

[lilcinw]

Athlon = Phenom II w/o L3 cache. It is based on the same arch.

As for Trinity, the supposed improvements put it ahead of or equal to the Phenom IIs. This is 2 modules versus 4 cores as well. I think this makes a full 4 module PD look quite promising.

However, this is AMD putting out the estimates and after getting burned by BD hype I am still skeptical.

 

[bawchicawawa]

Yeah, it has no l3 cache... That's why it isn't a P2 chip, I don't understand how you are arguing against my post?

 

[gamerk316]

Well, if you think that Llano nor Trinity are meant to be "power computing" solutions, I see no harm in them having less muscle than the "pure enthusiasts" CPUs.

But if the CPU side of the equation gets worse, Trinity becomes very hard to recommend. For its intended use [IE: Not gaming], you are more likely to notice a weak CPU over a weak GPU.

My argument is a simple one: If BD < PII, and Llano is based on Athlon, then its quite possible Trinity could be weaker then Llano, at least as far as the CPU goes.

 

[bawchicawawa]

But if the CPU side of the equation gets worse, Trinity becomes very hard to recommend. For its intended use [IE: Not gaming], you are more likely to notice a weak CPU over a weak GPU.

My argument is a simple one: If BD < PII, and Llano is based on Athlon, then its quite possible Trinity could be weaker then Llano, at least as far as the CPU goes.

You could argue that; but, first gen BD isn't THAT far from p2. With the current 2 updates BD did get a small 3% boost overall, as well, microsoft said that even with the updates, they don't represent the boost you will get with win8. Also, second gen BD will have improvements from the first gen, so your argument isn't really valid...

 

[noob2222]

http://news.softpedia.com/news/AMD-Moving-to-Tick-Tock-Micro-Architecture-Model-Some-Say-248736.shtml
not so interested

After actually reading the article, this is nothing like intel's tick-tock. AMD is going to adopt testing new sub-architectures with APU - no chache memory, half the core counts, followed by high end CPU with full caches and core counts. This will allow AMD to find any bugs and get them addressed hopefully between the APU and the CPU release.

The only time AMD will "tick" is dictated by the founaries they use, GF and TSMC mainly. Intel is forcing ticks, amd is adopting a tock-tock model.