AMD Trinity On The Desktop: A10, A8, And A6 Get Benchmarked!

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

Soul_keeper

Distinguished
Jan 23, 2009
321
17
18,815
As a llano owner it's hard to be impressed.
Not really an upgrade. Looking forward to the real desktop piledrivers.

I take it this will be the full lineup untill FM3 ...

This is a joke, make the APUs and FX chips share a socket.
Then people can actually upgrade something.
intel has it right.
 

agnickolov

Distinguished
Aug 10, 2006
520
0
18,980
Alas, the comparison I was really looking for just isn't in this review. There's clock for clock comparison with Bulldozer on page 2, but the Phenom II numbers are strangely missing. Throw in a Phenom II 975 for the same metric in the next review please...
 
G

Guest

Guest
I have a question here, can anyone give a FULL list of ATI Gfx cards that are hybrid-crossfire compatible?

I'm eyeing towards the HD 6770

 

boletus

Distinguished
Mar 19, 2010
69
0
18,630
Here's what is a bit amusing for me: when AMD bought out ATI, I would not have thought it would be ATI that would end up saving AMD's butt. But yeah, in retrospect, it was a very smart move.
 
G

Guest

Guest
I just hope what you started on page 2 - Piledriver/Bulldozer comparison - will get its own article. Comparing architecture differences.

At least what it looks like is that now AMD can compete with Phenom II X6 :)

And as agnicklov stated, hope to see more.
 

emad_ramlawi

Distinguished
Oct 13, 2011
242
0
18,760
Nice review , sad that the A10-5700 was not included , its okay not your fault though .

it was as expected really , small CPU improvement and big in GPU side , its sad that they will require a new socket thought , does any one know if the 2013 APU Kaveri will require a new socket as well , cause that will be lame and more like a laptop infrastructure than desktop , anyways thanks Tom Hardware for a good review , and those APU will make a good silent overall for anybody PC , anybody as in regular people with sata USB 3.0 and sata 6gb and good performance for the price
 
Great article Chris.

Can't wait till you do some benchies on how well it overclocks and if you can do much with the graphics in terms of overclocking too?

I'd like to know how well these go with a mid range discrete graphics card (and what series are compatable) in CF with the APU too?

Cheers !!

 

silverblue

Distinguished
Jul 22, 2009
1,199
4
19,285
So, Piledriver is easily faster than Bulldozer in single-threaded workloads... and the latter has L3 cache. That Vishera review can't come soon enough, really; with the cache, plus no iGPU hogging the memory bandwidth, AMD might finally have something that is, per core, generally faster than Phenom II. As always, Windows 8 testing might change scores slightly for all parties, so that's worth a glance as well.

The cache latencies may be unchanged, however the cache misses should be much reduced if AMD have done their homework. Still, the CPU might just be holding back the GPU; dropping that GPU core to Llano levels or overclocking the CPU to, say, 4.5GHz, will no doubt give us an answer on that one. Also, disabling one x86 core per module to see how it affects performance in comparison to Bulldozer will no doubt be an interesting test.
 

Edgar_Wibeau

Honorable
Jun 14, 2012
4
0
10,510
Regarding OpenCL Winzip:

Erm, 88 seconds versus 131 seconds: "32% improvement" - that's percentage calculation done the wrong way. It's 100*88/131 = 67%, so 33% (not 32%) less time, but looking at the IMPROVEMENT (calculations done in the same timeframe) of 100*131/88 = 149%, so the improvement ist 49%, not 32%.

Just my 63 THB.


- hope this post won't be double cause I had to register first and it didn't show up yet
 
G

Guest

Guest
I was in WalMart last week and saw an HP PC with a box that listed the A8-5500 as the CPU?
 

assafbt

Distinguished
Mar 25, 2011
26
0
18,530
Any benchmark that misses out on a "control-group" Low cost intel + low cost graphics, is a bit out in the air for me, even if the intel is better, I want to know how much faster and at what cost
 

cangelini

Contributing Editor
Editor
Jul 4, 2008
1,878
9
19,795
[citation][nom]Edgar_Wibeau[/nom]Regarding OpenCL Winzip:Erm, 88 seconds versus 131 seconds: "32% improvement" - that's percentage calculation done the wrong way. It's 100*88/131 = 67%, so 33% (not 32%) less time, but looking at the IMPROVEMENT (calculations done in the same timeframe) of 100*131/88 = 149%, so the improvement ist 49%, not 32%.Just my 63 THB.- hope this post won't be double cause I had to register first and it didn't show up yet[/citation]
Actually, I believe that you want the difference of the two times, which would be (131-88)/131, to get the improvement.
 
While I can clearly see that a lot of work went into this, I'm not sure it answers the relevant question, which is "Phenom II to Piledriver (and keep mobo and Windows license), or Phenom II to SB/IB (requiring a new mobo and new Windows license)?" You teased some references to the i3, so I was disappointed not to see it included in any of the benchmarks.
 

muy

Honorable
Feb 26, 2012
17
0
10,510
i'm an amd fanboy but i think it is pathetic that 2-3 year old amd chips still outperform amd's newest chips WHEN CLOCKED AT THE SAME SPEED. especially in games.

give me a phenom II on 32 nm please. i simply can't understand why amd is not doing that.
 

tomfreak

Distinguished
May 18, 2011
1,334
0
19,280
1. oh jeez AMD, what the heck are u thinking, why cant u redesign the whole line up of low-end HD7000 discrete GPU into VLIW4 with the same Shader count as APU IGP, to sync each other to use the traditional crossfire that work on most game title. Depending on Dual graphic now is a hit or miss thing, dual graphic 2 diff GPU architecture arent going to work better than crossfire 2 identical GPUs.

2. Please for god shake DO NOT put large L3 cache on bulldozer's successor piledriver. Add more x86 core probably do better now. Single threaded performance is important but whats more important is going head to head against Intel overall at least for multithreaded. I actually prefer 12 core piledriver to go head against 3770k than 8 core pildriver with L3 that doesnt perform good in single or multi-threaded place.
 
I am not as excited as most here are and yes there is improvement but enough sadly. It will be great for those who are stuck with anything worse (pokes at e-350 grrr) but at the same time it is a let down for those who already bought into Llano that had some hope that Trinity would be an all the way upgrade but when I had seen the clocks for the desktop models I knew that hope was pointless. So I will be sitting this round out and wait for the next gen parts next year >.<

Staying with my A8 3530mx and Phenom2 x4 for what looks like years to come.
 

gondor

Distinguished
Apr 14, 2011
80
14
18,635
[citation][nom]JiggerByte[/nom]So this means that a 'Crossfired' Trinity APU would beat ANY similarly-priced Intel (CPU+discrete GPU) ???Well at least in gaming[/citation]

Highly unlikely. Top Trinity chips (A10) are likely to cost as much as top of the line Llano (~$120, if not more), meaning that by saving $40-50 by going with cheaper dual-core Intel Sandy Bridge Pentium CPU and not wasting 1 GB of system RAM you can afford a HD7750 ($110) in place of HD6670 ($60) that you would use in Trinity Crossfire, and the combination would soundly demolish said Crossfire configuration in games, let alone single-threaded workloads which seem to be quite common in Tom's testsuite.

Crossfire configuration defies the very idea behind APUs. If you want to use discrete graphics to speed things up, go discrete all the way (plus you get much better scaling with upgrades).
 
G

Guest

Guest
Better GPU part (and resulting game perfomance increases) aside, the most interesting part, imho, is getting some first glimps of Piledriver cores.
With the addition of an (hopefully improved) L3 - I don't agree with Tomfreak, more cores won't give much benefit in anything besides highly multithreaded applications, which are still rare - they might finally be a replacement for Phenom IIs.

It seems that they might finally hit the clock rates necessary to get enough performance from this architecture , which, coupled with the IPC improvements might still not take them to the same level as Intel, but at least make them superior to the previous generation.

I never expected wonders from Piledriver, but it might at least be a decent enough upgrade to the underwhelming Bulldozer CPUs.
 

BSMonitor

Distinguished
Nov 19, 2007
167
0
18,680
Single core bulldozer with large L3 in a test where new data is consistently passing through the CPU... The L3 would almost certainly be harming the performance, not helping.. In a video compression app, I doubt a large cache helps. The app is not repeatedly looking at the same data locations.. I would guess there are a ton of cache misses in that test for bulldozer.
 
Here is a list of Win32 multi-threaded software began prior to adding cores to processors became a 'thing.' It's from one of the 2CPU bunch. It's an interesting look at how things really haven't changed a great deal . . . . :)

[citation][nom]muy[/nom]i'm an amd fanboy but i think it is pathetic that 2-3 year old amd chips still outperform amd's newest chips WHEN CLOCKED AT THE SAME SPEED. especially in games.give me a phenom II on 32 nm please. i simply can't understand why amd is not doing that.[/citation]

It's a dead-end and phantom performance gain, that's why.

AMD is in the 2nd-3rd stage of a process that unifies memory architecture and processes instruction-level C on what they call their SIMD Engine Array --> the GPU part of the APU.

From AMD's perspective, Trinity Piledriver is not a 15% improvement over Zambezi cores in performance; it's the next-to-last (next-to the next-to-last?) step to performance gains seen in multiples of 3X and 5x . . .

Bring on 28nm Kaveri with GCN cores (and HD7750 dual-graphics!) and let's see if OpenGL And OpenCL Can Overhaul Your Photo Editing Experience?



edit: Hope I fixed my links . . .




 

cknobman

Distinguished
May 2, 2006
1,117
263
19,660
@ the retards whining about "where are the Intel CPU comparisons":

This is a PREVIEW of pre production hardware with unfinalized BIOS and drivers. Chris was only trying to give us a PREVIEW of what to expect.

Once Trinity for desktop is officially released to the public channel you can guarantee there will be a full REVIEW including benches and comparisons to Intel CPU's.

Its kinda hard for Chris to give you a comparison with Intel parts when he has no price point to go off of.
 

Edgar_Wibeau

Honorable
Jun 14, 2012
4
0
10,510

That's what you're doing. But this is 33% less time. Which is not the same as an improvement of 33%. It's an improvement of 49%.

Let's be silly and talk money and bananas.

One day you're getting 10 bananas for 3 Dollars. Your Dollar performs at 3 1/3 Bananas.

Next Day you're getting 10 bananas for 2 Dollars. So that's 33% less Dollars for the same 10 bananas. One Dollar is now 5 bananas instead of 3 1/3. So your Dollars perform (improve) 33% better? Nope, they perform 50% better.

That's what we call an improvement. Same bananas, improved Dollar by 50%.

Same work (zipped content), same bananas, less Dollars (CPU time).

Got me?
 
Status
Not open for further replies.