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 ...

 

[triny]

At least you understand that instead of being smart and underestimating performance improvements Like a smart company should Amd Always overestimates their performance expectations, Like 50% more through-output when in reality its less then 20%. Amd really needs to run 50+ benchmarks of programs that people actually use and then find the price/performance ratio or does that make to much since!

That's odd I have always heard them claim that it was under certain load or benchmark when making claims like that
I guess some people only hear numbers the words don't register.
but if you have a link please share it

 

[triny]

This might explain some of the power savings in how they were able to get the higher clocks with Piledriver (FX-4120+GPU).

AMD packs Cyclos clock tech into Piledriver.

"To back its claims of power savings up, Cyclos has some real-world figures. Based on Piledriver-based x86 processing cores running at 4GHz and above, the resonant clock mesh technology drops clock distribution power by up to 24 per cent at peak and between five and 10 per cent on average in the company's testing. Clock-skew, a serious issue in high-speed processors, is claimed to be unaffected by the drop in power draw."


http://www.bit-tech.net/news/hardware/2012/02/21/amd-packs-cyclos-piledriver/1

good article

 

[triny]

haha, get out of the mirror...

tehehe
at least your funny some days and mal-mental all the time
see how I bring out your good side

 

[triny]

Intel can get away with overcharging Amd can't!

is that why 1000$ parts languish on the shelf ?
only an Intelidiot would buy chips for more than 400$
but you have a point there are a lot of Intelidiots

 

[triny]

PD and trinity are not mature products
We would have to wait for 2014 to see mature product with excavator
In the meantime trinity and piledriver will be an improvement over lano and bull doozer
a10 5800 over lano a8 3870
ipc+ clock+
a10 as a product will be vastly improved over the top lano
PD will be an improvement over Bull Doozer
I made my prediction

a10 5800 CPU 25-30% igp 60-80% compared to lano 3870
using +ipc and +clock speed

PD 10-15% per core over BD first iteration ( ipc and clock increase)

I find it odd that the people who question it are unwilling to make their prediction
known .Instead they act like an adolescent with their panties in a wad.

 

[Cazalan]

a10 5800 CPU 25-30% igp 60-80% compared to lano 3870
using +ipc and +clock speed

PD 10-15% per core over BD first iteration ( ipc and clock increase)

I find it odd that the people who question it are unwilling to make their prediction
known .Instead they act like an adolescent with their panties in a wad.

I have my doubts but the potential is there. I'd rather expect low and be pleasantly surprised than be disappointed again.

Trinity has a few benefits over bulldozer/llano:
1) About 6 months of process ramp up for GF. (Interlagos officially started shipping 9/7/2011)
2) IPC improvements from cache latency/scheduler fixes/improvements
3) GPU based on GCN (demonstrably better and not just from clock speeds)
4) Reduced power consumption from clock mesh. ("Clock tree power contributes nearly 40-45% of the total dynamic power in a chip.")
5) Higher CPU speeds
6) Higher memory speeds

Things are coming together on how they were able to do a 17W trinity.
What they'll be able to do in a 65W/100W profile remain to be seen.

 

[esrever]

Those bench's were done at 1920x1080 with 4xAF and 8xAF with a high end discrete GPU. That completely defeats the purpose of an APU to begin with. Half their die space is devoted to a GPU that your just disabling. Redo those bench's with the built in GPU vs the HD3K on the SB's and watch the APU's smoke them. Which is the entire point of the APU to begin with. You won't be doing high end gaming with a APU, it just won't work. It's good for mobile gaming and low power budget PC's. That was a good article for desktop decisions as it show's that a decent discrete GPU greatly diminishes the APU's worth.

You are aware the entire point of the article was to bench CPU's? All that was shown was that llano was hopelessly CPU bottlenecked compared even to $80 processors in most every major title, even when paired with the best GPU on the market.

My point being, llano will never be suitable for gaming with any significant graphical settings, simply because the CPU side is a cripple. Now theres nothing wrong for having a low power IGP for laptops/netbooks that is actually halfway decent, but as I've said before, I expect tablets/smartphones to make those two form factors obsolete within a decade. I simply don't see APU's as a major factor going forward.

the apu gpu is much much weaker than the cpu right now. I don't see your point.

 

[esrever]

@gamerk316: imo, this is exactly what 'they' refuse to understand. that article showed how a sub-$200 desktop cpu would perform for gaming. that's it. there was no mobile issue, no igpu issue. llano igpu's superiority was already demonstrated, but it's ability to drive a gaming gfx card was not demonstrated.
and who's to say what cpu should be tested or not? that's like ordering someone to abide by a restricted pc configuration (apple, anyone?).
when llano came out in june~ last year, i really, really wanted to build a llano pc. my plan was to use the igpu for a few months then add a radeon hd 6850 or gtx 560ti right around now and turn it into a cheap gaming pc. since the a8 3850s could be overclocked to 3+ ghz, i figured it's be great for a cheap (compared to other 'cheap' quadcore cpu available at that time - 2.8 ghz core i5 2300), single gfx card gaming pc. plus, i'd have a cheap 32 nm quad core cpu for multitasking and stuff. i knew that the apus were entry level, but they should at least be able to drive a mid range/upper mid range gfx card like cheap, entry level pentiums or slightly costlier core i3 cpus. i am a bit relieved that i didn't go that way. i still want a capable cpu + an igpu like llano's. llano set the bar for igpu, for desktop and mobile.

Llano was never designed to do that. You could have just bought a phenom 2 or i3 with a 6450 then upgraded.

 

[esrever]

At least you understand that instead of being smart and underestimating performance improvements Like a smart company should Amd Always overestimates their performance expectations, Like 50% more through-output when in reality its less then 20%. Amd really needs to run 50+ benchmarks of programs that people actually use and then find the price/performance ratio or does that make to much since!

when did they do this?

 

[fazers_on_stun]

so basically you and your partner there (triny) are going on a planning roadmap from a futures campaign conference for basis or foundation to your facts.?
really.?
did you also use the Bulldozer charts FROM AMD as motivation as well.?

http://i1130.photobucket.com/albums/m529/malmental/opteron20AMD20Excavator20architectu.gif

quote
" AMD representatives explained by factors which processor architecture the next few years will provide a performance gain with respect to its predecessors. This year Piledriver will bring improved frequency potential and an increase in the number of executed instructions per clock cycle. Steamroller in 2013 will increase performance in data parallelism. Finally, Excavator in 2014 will increase the productivity of some unknown application . Note that the image describes the evolution of server processors AMD, so not all development areas should be transferred to the desktop segment. "

end quote

Heh, the thing to note about AMD's PD, Steamroller & Excavator performance vs. time chart is that there are no numerical scales given. So that could be 0.001% performance increase over a timescale of 10 billion years 😛..

 

[de5_Roy]

Llano was never designed to do that. You could have just bought a phenom 2 or i3 with a 6450 then upgraded.

ph ii is old tech, 45 nm cpu, high load power consumption - reason why i skipped it (i didn't know back then that it'd game better than llano and fx... but high load power consumption was the main reason). also it lacked good on-board gpu support iirc. i skipped i3 because i needed a 4 core cpu. 6450 would add an unnecessary gfx card cost in the begining.
see, 'llano was never designed to do that' sounds like an excuse. athlon ii x2/x3's were designed to be entry level general purpose cpus but they were recommended as entry level gaming cpu. same with phenom ii x2, athlon ii x4 cpusand sandy bridge pentiums. now that apus are slowly replacing athlon iis and ph ii x2's, anyone buying entry level amd for a budget gaming machine will be forced to choose from them. not everyone chooses intel because it performs better, some people choose amd because it's amd. unfortunately they might not be getting same entry level cpu performance with llano as previous gen low end cpus.

 

[esrever]

ph ii is old tech, 45 nm cpu, high load power consumption - reason why i skipped it (i didn't know back then that it'd game better than llano and fx... but high load power consumption was the main reason). also it lacked good on-board gpu support iirc. i skipped i3 because i needed a 4 core cpu. 6450 would add an unnecessary gfx card cost in the begining.
see, 'llano was never designed to do that' sounds like an excuse. athlon ii x2/x3's were designed to be entry level general purpose cpus but they were recommended as entry level gaming cpu. same with phenom ii x2, athlon ii x4 cpusand sandy bridge pentiums. now that apus are slowly replacing athlon iis and ph ii x2's, anyone buying entry level amd for a budget gaming machine will be forced to choose from them. not everyone chooses intel because it performs better, some people choose amd because it's amd. unfortunately they might not be getting same entry level cpu performance with llano as previous gen low end cpus.

I would think very few people buying llano will think about upgrade it beyond adding a 6670. The chip works well as an all in 1 but beyond that, you are just losing half the die area to gpu you are wasting. The cpu isn't really that bad, its about 5% faster than the athlons and quadcore if you need a quad core. generally speaking, If die area is used for something and its going to waste, its limiting the chip for the same power/performance/price envelope.

Your example was very specific and you could not find an satisfactory solution not because anything at fault with llano but with the extensive criteria that could not be met by any cpu. You just wanted too much from something that could not be made.

 

[triny]

I have my doubts but the potential is there. I'd rather expect low and be pleasantly surprised than be disappointed again.


Things are coming together on how they were able to do a 17W trinity.
What they'll be able to do in a 65W/100W profile remain to be seen.

The clock mesh is good idea all around for Intel as well
but it needn't be restricted to 17watt solutions
it will work on 100 watt parts as well to good effect enough to make a10 5800 125watt need only 100 watt to do the same

I try not to expect anything and be impartial

 

[triny]

I would think very few people buying llano will think about upgrade it beyond adding a 6670. The chip works well as an all in 1 but beyond that, you are just losing half the die area to gpu you are wasting. The cpu isn't really that bad, its about 5% faster than the athlons and quadcore if you need a quad core. generally speaking, If die area is used for something and its going to waste, its limiting the chip for the same power/performance/price envelope.

Your example was very specific and you could not find an satisfactory solution not because anything at fault with llano but with the extensive criteria that could not be met by any cpu. You just wanted too much from something that could not be made.

They used Opteron to streamline their lineup I would imagine they will allow the desktop apu to use any 7000 series cards.whether that is with trinity or Kaveri or later is undetermined.
but the HSA aspect is what really intrigues me.That up 113% is ominous
Plus the clock mesh .

 

[triny]

Seriously, dont you guys just wished PD and trinity would be a big Flop so AMD would go bankrupt and Samsung will come in and buy them out, thus making them competitive again ?
at least on the R&D budget and Fab funds front ?

I can see it now
Samsung home entertainment One stop shop

 

[esrever]

Seriously, dont you guys just wished PD and trinity would be a big Flop so AMD would go bankrupt and Samsung will come in and buy them out, thus making them competitive again ?
at least on the R&D budget and Fab funds front ?

Why would samsung want to buy AMD?

 

[-Fran-]

Seriously, dont you guys just wished PD and trinity would be a big Flop so AMD would go bankrupt and Samsung will come in and buy them out, thus making them competitive again ?
at least on the R&D budget and Fab funds front ?

AFAIK they'd lose the x86 licensing from Intel. AMD has to be AMD until it dissapears or Intel decides to licence x86 to someone else.

And we're going in loops here, lol. Come on AMD, you're making us lose our sane minds here!

Cheers!

 

[Cazalan]

Seriously, dont you guys just wished PD and trinity would be a big Flop so AMD would go bankrupt and Samsung will come in and buy them out, thus making them competitive again ?
at least on the R&D budget and Fab funds front ?

I wouldn't really want to see a Korean company buying a pioneering USA company like AMD. Intel sure.

 

[Cazalan]

I can see it now
Samsung home entertainment One stop shop

They pretty much are a 1 stop shop.

From building construction, to machinery, hotels, resorts, insurance, LCD, to cell phones, tablets, etc.

 

[noob2222]

Seriously, dont you guys just wished PD and trinity would be a big Flop so AMD would go bankrupt and Samsung will come in and buy them out, thus making them competitive again ?
at least on the R&D budget and Fab funds front ?

I don't think anyone would opt to buy AMD, whoever did would be looking to get ass raped by Intel, lawsuits, bribery, and outright dirty tactics. Who wants to compete with that.

No, if AMD goes under, we will only see the biggest corrupt monopoly in history. Prices would double and development would come to a screeching halt. Intel would just sit back and rake in the cash.

But keep wishing for AMD to go under, we may see how it plays out. Intel laughed at the fines by the EU, 5 years of profits, 30B, fined 1B for breaking the law, pretty good trade off.

 

[Chad Boga]

I don't think anyone would opt to buy AMD, whoever did would be looking to get ass raped by Intel, lawsuits, bribery, and outright dirty tactics. Who wants to compete with that.

No, if AMD goes under, we will only see the biggest corrupt monopoly in history.

In history? WOW!!! :lol:

Prices would double and development would come to a screeching halt. Intel would just sit back and rake in the cash.

Pricing would surely rise somewhat in the short term, but thankfully there is the ARM threat to keep innovation ticking along.

 

[nforce4max]

I have my doubts but the potential is there. I'd rather expect low and be pleasantly surprised than be disappointed again.

Trinity has a few benefits over bulldozer/llano:
1) About 6 months of process ramp up for GF. (Interlagos officially started shipping 9/7/2011)
2) IPC improvements from cache latency/scheduler fixes/improvements
3) GPU based on GCN (demonstrably better and not just from clock speeds)
4) Reduced power consumption from clock mesh. ("Clock tree power contributes nearly 40-45% of the total dynamic power in a chip.")
5) Higher CPU speeds
6) Higher memory speeds

Things are coming together on how they were able to do a 17W trinity.
What they'll be able to do in a 65W/100W profile remain to be seen.

Trinity isn't GCN but older VLW4.

 

[nforce4max]

the apu gpu is much much weaker than the cpu right now. I don't see your point.

Do you own any AMD APUs and got physical proof to back that up? In my personal experience it is the other way around and from your past posts your thinking some times is often backward. Sure the cpu side can't hold up much but the gpu isn't garbage either. Llano shines in the mobile segment where alternatives are limited. The limited IGP performance that you are seeing is mainly due to low clocks and the well known memory I/O limitations that are fixed in Trinity.

 

[nforce4max]

Seriously, dont you guys just wished PD and trinity would be a big Flop so AMD would go bankrupt and Samsung will come in and buy them out, thus making them competitive again ?
at least on the R&D budget and Fab funds front ?

I hope AMD doesn't go under, if they do we're in a lot of trouble as it is one of the worst things that can happen to the industry.

 

[esrever]

Do you own any AMD APUs and got physical proof to back that up? In my personal experience it is the other way around and from your past posts your thinking some times is often backward. Sure the cpu side can't hold up much but the gpu isn't garbage either. Llano shines in the mobile segment where alternatives are limited. The limited IGP performance that you are seeing is mainly due to low clocks and the well known memory I/O limitations that are fixed in Trinity.

the cpu is a athlon x4 and the gpu is a 6550. thats a cpu heavy system in any game. You'd want a 6770 with an athlon x4 to be balanced imo.