AMD Piledriver rumours ... and expert conjecture

Page 45 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
Reynod

Reynod

Administrator
Mar 26, 2007
14,921
16
42,215
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 ...
 
Sort by date Sort by votes
@triny: trinity might not have hsa. it will use older vliw arch. iirc hsa is gcn's successor.
higher clocked cpu/apus usually have 100 mhz clockrate increase. so a8 -> a10 difference might be only 100 mhz.
piledriver is (according to amd)improvement of bulldozer architecture.
your cpu/gpu improvement figures are a bit overenthusiastic. i'd suggest you to keep your expectations lowered, in case anything happens.
@yuka: thanks for the salt! :)
edit: sorry for the double negative. 😛
 
Upvote 0 Downvote
- ...Slides as the bottleneck style of guitar playing.”

BD (The Flying Tailor FT-8150 LCS WOX $398.29) hoped would be a parachute of course this was not the original objective,
about the extreme waterproof bench are on the way.
 
Upvote 0 Downvote
@triny: trinity might not have hsa. it will use older vliw arch. iirc hsa is gcn's successor.
higher clocked cpu/apus usually have 100 mhz clockrate increase. so a8 -> a10 difference might be only 100 mhz.
piledriver is (according to amd)improvement of bulldozer architecture.
your cpu/gpu improvement figures are a bit overenthusiastic. i'd suggest you to keep your expectations lowered, in case anything happens.
@yuka: thanks for the salt! :)
edit: sorry for the double negative. 😛
Click to expand...

My prediction was based on AMD literature not my expectations so they may well be based on AMD hype
Though I doubt IBM would get involved based on hype .comparing a8 lano 2.9ghz and 3.8ghz a10 trinity also considering hd7670
will be more robust than hd6670
would you care to make a prediction?
 
Upvote 0 Downvote
@triny: trinity might not have hsa. it will use older vliw arch. iirc hsa is gcn's successor.
higher clocked cpu/apus usually have 100 mhz clockrate increase. so a8 -> a10 difference might be only 100 mhz.
piledriver is (according to amd)improvement of bulldozer architecture.
your cpu/gpu improvement figures are a bit overenthusiastic. i'd suggest you to keep your expectations lowered, in case anything happens.
@yuka: thanks for the salt! :)
edit: sorry for the double negative. 😛
Click to expand...

This is what I have been saying.....

My prediction was based on AMD literature not my expectations so they may well be based on AMD hype
Though I doubt IBM would get involved based on hype .comparing a8 lano 2.9ghz and 3.8ghz a10 trinity also considering hd7670
will be more robust than hd6670
would you care to make a prediction?
Click to expand...

I love the almost 1GHz clock speed difference. Thats a cloock speed difference of about 31% right there. They are saying about 20-30% in CPU performance improvements? If so that means Trinitys CPU itself (based on PD) only has better performance because of the 31% clock speed increase, not because the arch is better.

If its only 20% better CPU wise than Llano, then that means that Trinitys CPU would be about 10% less in performance.

Better hope thats not the comparison they are making or else it doesn't look too good for AMD. If so they are pulling a NetBurst, trading performance for higher clock speed.
 
Upvote 0 Downvote
My prediction was based on AMD literature not my expectations so they may well be based on AMD hype
Though I doubt IBM would get involved based on hype .comparing a8 lano 2.9ghz and 3.8ghz a10 trinity also considering hd7670
will be more robust than hd6670
would you care to make a prediction?
Click to expand...
amd's (or intel's) ghz numbers means a little or nothing. architecture, efficiency, real world performance are better for determining cpu/apu's worth.
afaik, discreet 7670 will be a rebadged 6670 (which is a minor improvement of 5670). gpus from 77xx series and higher have gcn. trinity's integrated gpu will have vliw4 (from 6900 series) and some minor stuff from the 7000 series.
i really don't like predicting...so many things can go wrong...
anyway..... i will give it a shot.
from what i've seen and read, trinity seems quite potent. price, gpu performance are in favor of trinity. amd's current success with their apus is also a big plus.
but, trinity's biggest obstacle is amd itself. amd sorta crippled llano. they intentionally clocked it's cpu lower. even overclocked llano doesn't deliver high cpu performance. the gpu can't speed down afaik (intel's igpu has turbo boost). desktop llanos run hotter because of the gpu. apus do not support amd's full instruction set like their cpus do (compare zambezi's and llano's supported instructions). so far, quick sync has more app support than amd's vce. ivb will have quick sync 2.0 and more app support.
imo having two unsuccessful fab-partners is worse than having one. glofo has already cost amd.
amd's imc (integrated memory controller) is inferior compared to intel's. an improved imc can bring out very high performance from the apus.
based on these, i'd say trinity will find moderate sucess in general purpose laptops. if the yields are bad, they'll be stuck there instead of breaking into ultrathins...might cost amd another apple contract...
if amd and it's partners can deliver enough apus, amd will undercut intel. intel's decision to not lower cpu prices (i've been saying that for a while now 😛) for ultrabooks will backfire on them...big time. if they run into a chipset bug like cougar point bug, they'll face even more problems.
cpu performance (not clockspeed) will roughly improve 7-10% per [strike]core[/strike] module, gpu performance improve 30-60% depending on apps and games. power efficiency will be the deciding factor for success in laptop arena.
edit: thought i should add this poll from dailytech.
 
Upvote 0 Downvote
This is what I have been saying.....



I love the almost 1GHz clock speed difference. Thats a cloock speed difference of about 31% right there. They are saying about 20-30% in CPU performance improvements? If so that means Trinitys CPU itself (based on PD) only has better performance because of the 31% clock speed increase, not because the arch is better.

If its only 20% better CPU wise than Llano, then that means that Trinitys CPU would be about 10% less in performance.

Better hope thats not the comparison they are making or else it doesn't look too good for AMD. If so they are pulling a NetBurst, trading performance for higher clock speed.
Click to expand...

They are actually saying that trinity 35 watt will perform same as lano 65watt. what ever that means
a8 to a8 100 tdp they say will be +25% cpu +50% igp
an A10 125 tdp is gravy on top
going by AMD numbers


 
Upvote 0 Downvote
amd's (or intel's) ghz numbers means a little or nothing. architecture, efficiency, real world performance are better for determining cpu/apu's worth.
afaik, discreet 7670 will be a rebadged 6670 (which is a minor improvement of 5670). gpus from 77xx series and higher have gcn. trinity's integrated gpu will have vliw4 (from 6900 series) and some minor stuff from the 7000 series.
i really don't like predicting...so many things can go wrong...
anyway..... i will give it a shot.
from what i've seen and read, trinity seems quite potent. price, gpu performance are in favor of trinity. amd's current success with their apus is also a big plus.
but, trinity's biggest obstacle is amd itself. amd sorta crippled llano. they intentionally clocked it's cpu lower. even overclocked llano doesn't deliver high cpu performance. the gpu can't speed down afaik (intel's igpu has turbo boost). desktop llanos run hotter because of the gpu. apus do not support amd's full instruction set like their cpus do (compare zambezi's and llano's supported instructions). so far, quick sync has more app support than amd's vce. ivb will have quick sync 2.0 and more app support.
imo having two unsuccessful fab-partners is worse than having one. glofo has already cost amd.
amd's imc (integrated memory controller) is inferior compared to intel's. an improved imc can bring out very high performance from the apus.
based on these, i'd say trinity will find moderate sucess in general purpose laptops. if the yields are bad, they'll be stuck there instead of breaking into ultrathins...might cost amd another apple contract...
if amd and it's partners can deliver enough apus, amd will undercut intel. intel's decision to not lower cpu prices (i've been saying that for a while now 😛) for ultrabooks will backfire on them...big time. if they run into a chipset bug like cougar point bug, they'll face even more problems.
cpu performance (not clockspeed) will roughly improve 7-10% per [strike]core[/strike] module, gpu performance improve 30-60% depending on apps and games. power efficiency will be the deciding factor for success in laptop arena.
edit: thought i should add this poll from dailytech.
Click to expand...


As per AMD 7670 will be oem only seeing that they wont be using 6670
I am guesing hd7770
 
Upvote 0 Downvote
As per AMD 7670 will be oem only seeing that they wont be using 6670
I am guesing hd7770
Click to expand...
afaik, 7770 won't be able to use hybrid crossfire with trinity (vliw4) because it has different underlying architecture (gcn). besides, i don't think amd would allow relatively higher end 77xx series gpus to crossfire with apus. they don't allow 67xx series cfx with llano.
the specs you posted earlier looked very very similar to 6670.
 
Upvote 0 Downvote
afaik, 7770 won't be able to use hybrid crossfire with trinity (vliw4) because it has different underlying architecture (gcn). besides, i don't think amd would allow relatively higher end 77xx series gpus to crossfire with apus. they don't allow 67xx series cfx with llano.
the specs you posted earlier looked very very similar to 6670.
Click to expand...


That's a good point

I don't think they would use 6670 though they might until 2013
 
Upvote 0 Downvote
I think you need to just slow down and let things fall into place.
besides you have nothing more important to add at this time.
juts re-wording something you already said like three pages ago and haven't stopped yet.
starting to look like a troll.
just saying.
Click to expand...

+1

If BD taught me anything it is this: Expected performance and actual performance can be wildly different things.
 
Upvote 0 Downvote
This slide agrees with that for 17W Trinity vs 35W Llano

http://semiaccurate.com/assets/uploads/2012/02/Trinity-Improvements.png

Double performance/watt at roughly 1/2 the power = same performance

So, what is the performance of a 35W Llano?
Click to expand...

The Quality for each element on the left side of the equation. It is very important.
Example: Llano Which 35 W x2 cores 1.9 GHz HD 6480G 35 W. and 1333 DDR.
RAM performance has a significant effect on this one, these facts are well known.
Controversial use of poor models that's not a new idea. Llano is plain, not so advanced as they think.
 
Upvote 0 Downvote
The Quality for each element on the left side of the equation. It is very important.
Example: Llano Which 35 W x2 cores 1.9 GHz HD 6480G 35 W. and 1333 DDR.
RAM performance has a significant effect on this one, these facts are well known.
Controversial use of poor models that's not a new idea. Llano is plain, not so advanced as they think.
Click to expand...

I'm having a hard time understanding what you're trying to say here, but I think you raise a good point.

Does this performance/watt increase include faster memory speeds on Trinity or this assuming everything is the same except the APU?

In addition, assuming 2 core Llano vs 2 module Trinity is this with a parallelized workload?

This slide doesn't really give any useful information. Why does AMD think it's acceptable to present a bunch of meaingless charts showing their chips' superiority with little to no proof that the graphs are meaningful to common workloads.
 
Upvote 0 Downvote
I'm having a hard time understanding what you're trying to say here, but I think you raise a good point.

Does this performance/watt increase include faster memory speeds on Trinity or this assuming everything is the same except the APU?

In addition, assuming 2 core Llano vs 2 module Trinity is this with a parallelized workload?

This slide doesn't really give any useful information. Why does AMD think it's acceptable to present a bunch of meaingless charts showing their chips' superiority with little to no proof that the graphs are meaningful to common workloads.
Click to expand...
Mr. Akrou AMD- “Since you asked twice we are going to get you an answer.”

Which 2 models are used for main purposes?

left side: Llano
right side:Trinity the High Performance model + A better RAM

http://semiaccurate.com/assets/uploads/2012/02/Trinity-Improvements.png
 
Upvote 0 Downvote
Upvote 0 Downvote
Trinity supposedly 25% better cpu wise than Llano. A little bit of math using Tom's Adobe Premiere Pro benchmark:

Llano=69.2 minutes. -25%=51.9 minutes, or still under preforming a i3-2100(47:11).

That still doesn't look that good for a trinity A8, the A10 will be even better, but it would still not even come close to a 2500k(24:47).

On the graphics side, 50% is a huge jump, and would completely smash anything Intel will offer with ivy, even if Intel get's their 50% better than SB.

I highly doubt either side will get 50% better on graphics, but I think that 20-25% on trinity cpu will be possible.

TBH, This A8 vs i3 was pretty sad for AMD, considering they are marketing trinity at the high end of the mainstream market.
 
Upvote 0 Downvote
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom