Battle At $140: Can An APU Beat An Intel CPU And Add-In Graphics?

Page 3 - 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.
[citation][nom]ivaroeines[/nom]On the last AMD vs Intel test i wrote a piece that accused Toms hardware of being biased against AMD, to me this article/test manage to confirm my suspicions. Here are the things that make me think that Toms hardware is biased. 1. Toms indicates that the Intel motherboards are cheaper than AMD ones while on the test they use a Intel motherboard that is more expensive than the AMD one, this is normally no real problem. The problem is that they should use motherboards that have roughly the same specs, the cheap intel motherboards listed on newegg only support DDR3 1333 ram, if you want a Intel motherboard that support DDR3 1600 which all AMD boards supports you have to go up in price by 70%. 2. The power usage benchmark shows that the Intel system uses 107 watts fully loaded, if you count in the fact that a HD 6670 alone uses around 65 W fully loaded the rest of the system uses around 42 W, this with a cpu with a tdp of 65 W, i find that strange. My conclusion here is that either Toms changed the cpu during the test or they use a different cpu than the listed one. My guess is that they used the G620T ( like some have indicated before me ) in this test, well this is most likely a slower cpu than the G620 but dont know if its slower, but it is more expensive than the G620. Either way it sets some question marks on the whole test. 3. Toms claim that the G620/HD 6670 is $10 cheaper than the A8 3870K, my own search on newegg says that the price difference is less than a $1.As a end note, i dont dispute the fact that Intel makes better cpu's than AMD, they do, but Toms hardware need to take a look at how they conduct their tests, im starting to question whether i should trust them any more.[/citation]

Very very good opinion, totally agree!
 
im sorry.. but it just occurs to me that sandybridge dont scales well with memory speed right?

there is no significant difference of using 1333 or 1600 memory on intel platform compare to AMD APU platform...

i am not biased toward Intel but in this article what i expected to see is the result of using the same mobo and ram ( in prices ) + intel/dscreet / AMD APU.. guess toms just blows it for me when it uses more powerful platform on Intel... 🙁

http://www.anandtech.com/show/4503/sandy-bridge-memory-scaling-choosing-the-best-ddr3/6

 
[citation][nom]beavermml[/nom]im sorry.. but it just occurs to me that sandybridge dont scales well with memory speed right? there is no significant difference of using 1333 or 1600 memory on intel platform compare to AMD APU platform...i am not biased toward Intel but in this article what i expected to see is the result of using the same mobo and ram ( in prices ) + intel/dscreet / AMD APU.. guess toms just blows it for me when it uses more powerful platform on Intel... http://www.anandtech.com/show/4503 [...] est-ddr3/6[/citation]

They can use the same money for hall pack, but they didn, first of all. And also linked article is about I7 2600k, which is one of the high-end and fastest processors. And because shared benches are in gaming, where bottleneck is GPU, and not this CPU, this is not linked to the theme of our conversation again 😀 :)

Only i want to see really professional review, with proffessional approach in given architectures. This article has not the professional approach, but instead has incompetent approach to AMD Liano platform.
 
Hey Toms, do you think you can make a quick review with the A8-3870K APU and the Radeon HD 6670 video card in a hybrid crossfire configuration and benchmark it?

I am really interested in what the results would be, and no other publication has even tried to benchmark this type of configuration. The HD 6670 is the highest discrete card that can crossfire with an HD 6550D. Would it be a good value performance wise to have this configuration?
 
The reason why Llano responds to higher clocked RAM is because like someone said earlier, the GPU relies on system RAM. So IMO, for an APU 1866mhz and above should be the standard for all builds to get the most out of the chip.
 
[citation][nom]ubercake[/nom]The good thing to see is the progress AMD is making with their APUs. It would be nice to get to a point where you can have really good gaming-level 1080p video performance playing on high or ultra settings at 60Hz without the added expense, heat or power consumption of discrete video solutions. I'd pay double or triple the $140 for something like that.[/citation]

Isn't this what Piledriver is going to try and accomplish? But I agree, get the fusion chips to a point where it at least delivers 30-60fps at 1080p and we've got a winner. Less heat, power consumption, noise, and no discrete gpu then shove it into the tiniest form factor possible. I would build an HTPC off that in a heart beat.
 
[citation][nom]BSMonitor[/nom]And then there is Ivy Bridge... And you will see HD4000 and quad-core Core i5 for $140.. Oops, game over.[/citation]

Will they demo a real game this time? lol

Anyway, I'd like to see an i5 for $140... Don't know where in the world you get the idea that Intel will price it anything below $180 like the Sandy Bridge i5-2300.
 
[citation][nom]peroludiarom[/nom]I will say that its really unproffessional to use 1600mhz DDR3 with Liano, while manufacturer says its support 1866mhz! I recently got 3870K for my frend, and not even 1866 will show much bettter numbers in CPU and GPU, but 1866mhz even bottleneck the APU when you overclock. For best results i was using 2000mhz ddr3 ram.Also i'm sick to watch so poor ddr3 bandwidth performance on AMD overclocked processors here! Please start to overclock the memory controller - the whole world already knows this, but not Toms! With PII 555(x4) i got 16,5g/s in Sandra for my everyday overclock, and you show 12-14gb/s for PII 980?? Also using cheaper MB is already know that bottleneck Overclocking and even standart performace!Please Toms, bench again with right settings![/citation]

Why would a budget system (cause that is obviously the focus of this article) use a more expensive RAM than 1600?
 


Since Intel has convinced the market that FX is crap compared to theirs, they are going to price i5 like 250$ for sure
 

Then why the hell are Tom using a much more expensive motherboard for Intel?
If they want a true comparison they should not do that.
I seriously think Tom should redo this test.
So many doubts about Ram, motherboard, Power consumption.
Please think about it tom :non:
 
The power consumption table was quite alarming.
I thought the APU would have some advantage there but it used 50% more power at load!
That is horrendous for netbook/laptop space.
 
To someone with "X" dollars to spend, fairness is irrelevant. I don't care how fast an A8 would be with DDR3-2133 if the best RAM the budget buyer can afford is DDR3-1333. CrossfireX results are a mixed bag. I'd like to see an A8 with a HD6570 or HD6450, but along with those results needs to be a "reminder table" of games that came out before there were working Crossfire drivers (e.g. BF3 and Skyrim) or games that don't support Crossfire at all.
 
Very nice comparison. I think there are many people who wondered which setup would be better on a tight budget. It goes to show that Llano is not a fitting component in desktops, whatever the case. Add on the extreme upgrade ability that LGA 1155 holds, and there is no contest.
 
[citation][nom]saturnus[/nom]From the power consumption figures it's obvious that it is indeed the G620T they have tested but stated the price for the G620. For shame Tom's.[/citation]

We purchased a retail G620 from Newegg, it's not a 620T.

The 35 W figure was a typo. Fixed!
 
[citation][nom]mitunchidamparam[/nom]Then why the hell are Tom using a much more expensive motherboard for Intel?[/citation]

We address this in the test setup and benckmark page, guys.

We've proven time and time again that more expensive motherboards don't improve performance, they only give more features and in some cases more overclocking headroom. But we didn't overclock the Pentium, did we? We didn't even overclock the discrete Radeon, but we could have.

1155 boards start at $50 and FM-1 boards start at $60 on Newegg right now. Any complaints about motherboard cost are irrelevant.
 
[citation][nom]Cleeve[/nom]We address this in the test setup and benckmark page, guys.We've proven time and time again that more expensive motherboards don't improve performance, they only give more features and in some cases more overclocking headroom. But we didn't overclock the Pentium, did we?1155 boards start at $50 and FM-1 boards start at $60 on Newegg right now. Any complaints about motherboard cost are irrelevant.[/citation]
As one guy say before, you use more expensive MB, because if you dont, you cant use DDR3 1600... And Cleeve, please answer the question about RAM used, and setting in overclocked mode, where memory controller is also missed to tweak. Thanks
 
I just recently got a 550ti for my B-day and swapped my Wifes g620/onboard to push it vs my a6-3500 pushing the 550ti so she could have some graphics ponies as well. It was a world of difference pushing the same Graphics card both processors/apu $70-79 when purchased so it is about as direct comparison as it gets honestly. APU was around 12k mark 3dmark06 with 550ti (ghost multi 4.7 actualy I believe 3.0ghz) and stock 2.6ghz g620 was 14k marks 3dmark06 with 550ti. Game wise GPU has opened up another 10-20 fps 1080p most game mid or high not ultra. So if your going descrete card always Intel. If your going on a real budget honestly skip the a8's and live with a a6 or go intel. Sad part is most people with real budgets end up buying a big box pc and never get the max performance per dollar in tight situations.
 
ow and FM-1 MB start at $55, and that one comes with a $10 rebate taking the cost to $45(if purchased by the end of today), allowing for DDR3 1866 out of the box.
 
Oh yes forgot in USA both mother boards were cheap grade $50-60 ECS or BIOWARE with 8gb of Gskill 1600mhz 9-9-9-24 for both so budget nearly identical with generic 7200rpm hard drives no black editions or hybrids.
 


So you guys want motherboard and RAM included in the cost comparison?
And you want me to do the best I can in the same price total for both sides?

If the rest of the platform gets included in the cost you're shooting the AMD system in the foot. If your argument is that Llano *needs* 8 GB of superfast ram to make a decent budget system, then we should optimize the Intel system for cost, too. That means cheap RAM, and 4 GB will do because the graphics card has its own dedicated graphics RAM on the card.

Here's the realistic scenario:

OPTIMIZED AMD SYSTEM:
-------------------------------
Cheapest FM1 board: ECS A55F-M3 - $55
Cheapest 8GB 1866 RAM: Team Xtreem LV 8GB (2 x 4GB) - $56
A8-3870K: $140
-----------------
$251

(actually, this might be underpriced because I believe FM1 might require the more expensive A75 chipset for dual-channel 1866 RAM. So the board cost would go up another $30. But for now, let's give AMD the benefit of the doubt)

OPTIMIZED INTEL SYSTEM:
---------------------------------
Cheapest 1155 board: Jetway GTI61AG3: - $50
Cheapest 4GB 1600 RAM: Crucial Ballistix Smart Tracer 4GB (2 x 2GB) - $25
Pentium G620 - $70
HIS H577FL1GD Radeon HD 5770 1GB 128-bit GDDR5 - $105
-----------------
$250


So do you guys really want me to go ahead and build this comparo? I don't know what you think it will prove, because app results will be the same and Llano's graphics will get an insignificant boon from the faster memory, while the pentium system will get a colossal graphics boost from the 5770.

Or do you suggest I completely optimize the AMD system and then cripple the Intel's by not making use of it's strengths when it comes to cost savings?

DUMB INTEL SYSTEM:
--------------------------
Cheapest 1155 board: Jetway GTI61AG3: - $50
Cheapest 8GB 1866 RAM: Team Xtreem LV 8GB (2 x 4GB) - $56
Pentium G620 - $70
HIS H667FS1G Radeon HD 6670 1GB 128-bit DDR3 - $75
-----------------
$251

But who in his right mind would choose this over the first option? And the sad part is, *Intel results won't change one iota from the original test because motherboard cost does not impact performance*. If anything it'll probably increase a bit in apps, because despite the fact that Intel doesn't scale as quick with faster memory it *will* scale a little.

On a side note, nobody seems to be commenting how AMD got the benefit of the doubt BECAUSE I NEVER OVERCLOCKED THE DISCRETE CARD ON THE INTEL SYSTEM, by the way.

So to be fair, shouldn't I overclock the discrete Radeon also?

Or do you guys want me to avoid that? Because it seems to me you're making every little nitpick you can about the AMD system while you're happy to leave the Intel system with unrealistic handicaps.
 
Is this article about Intel has to have an AMD video card or can an overclocked A8 compare to a 6670? If its to say anything about the CPU's you would need an even GPU to the A8's to compare. I think that is a 6550 not 6670. If the A8's GPU part is overclocked maybe a 6570 would make a good compare.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom