The Brazos Review: AMD's E-350 Supplants ION for mini-ITX
by Anand Lal Shimpi on January 27, 2011 6:08 PM ESTPower Consumption: Better than Atom
Power efficiency was a big draw of Atom, but does AMD sacrifice any of that in order to deliver the performance it does with the E-350? To be blunt: no, not at all.
I don’t have any pico PSUs or anything super efficient readily available so don’t expect any of the numbers to be particularly impressive, but what they are is comparable to one another. I hooked up each one of the systems I’d been using to the same PSU and measured power in three conditions: idle, full CPU load (Cinebench 11.5) and while playing a 1080p H.264 video.
Pine Trail and the old ION platform consume just about the same amount of power at idle. The Athlon II system obviously draws more, in this case an increase of 17%. The E-350 uses less than 70% of the power of the Atom D510 system at idle.
Under load the Brazos advantage shrinks a bit but it’s still much lower power than Atom. While playing a H.264 you’re looking at ~83% of the power of an ION system, and 85% under full CPU load.
Say what you will about Intel’s manufacturing process advantage, it’s simply not put to use here with Atom. AMD’s E-350 is higher performing and uses less power than Intel’s 45nm Atom D510. Did I mention it’s built on a smaller die as well?
I wanted to isolate the CP...err APU and look at its power draw exclusively. I ran the same three tests but this time I’m not measuring power at the wall, but rather just power over the ATX12V connector directly to the CPU.
At idle the E-350 APU only requires around 3W of power. That’s actually not as low as I’d expect, especially given that Sandy Bridge is typically down at 4W when fully idle. AMD is apparently not being too aggressive with stopping clocks and gating when fully idle, at least on the desktop Brazos parts.
| Power Consumption Comparison | |||||
| ATX12V Power Draw | Idle | 1080p H.264 Decode | Cinebench 11.5 | ||
| AMD E-350 | 3W | 8W | 9W | ||
| AMD Athlon II X2 255 | 7W | 12W | 47W | ||
Under load, either full CPU or when using the video decode engine, APU power consumption is around 8 - 9W. By comparison, an Athlon II X2 255 will use 12W when decoding video (this doesn’t include the UVD engine in the 890GX doing most of the heavy lifting. The more interesting comparison is what happens when the CPU cores are fully loaded. The E-350 uses 9W running Cinebench 11.5 compared to 47W by the Athlon II X2.
Facebook
Twitter
RSS
176 Comments
View All Comments
Belard - Tuesday, February 1, 2011 - link
It would fail at bluRay 3D.the 1080i is a bit unimportant since flatscreen TVs don't do interlace.
doctorpink - Tuesday, February 1, 2011 - link
if i remember correctly, nano runs 1.8ghz and is much faster but more power hungry...are there netbooks with nano anyway?
grego3d - Tuesday, February 1, 2011 - link
I'm sold on this Sandy Bridge / Atom killer. Well, maybe not Sandy Bridge killer yet, but i'm sure the second generation fusion processors will be. Now I need help finding a great little mini-itx case. I'd love to build a mini pc the size of my Wii. I have found a few small cases, but where on earth are the slot feed DVD/BD players. I always start my daily reading right here @ Anandtech.com so please save me some time and help us all out by rounding up a case review for this new Fusion platform. Go AMD Fusion - Boo Intel (and your 1$Billion oops!)Belard - Tuesday, February 1, 2011 - link
Sure is such excitement over such a tiny chip.It also reminds me why one of the P4s Dells I use at work is sooooooooooo Slooooooooooooooow.
I'll admit I'm using my very old computer (AMD X2 3800) as an HTPC somewhat, since it wasn't worth selling when I upgraded. But its a 90watt CPU... so saving power is the big thing, eh?
lovansoft - Tuesday, February 1, 2011 - link
I've got an Acer Aspire 5517 with an AMD Athlon tk-42 processor and integrated HD3200 video. It's also 1.6 Ghz so I wanted to showcase a relative clock for clock comparison. It is a 20W chip with 1meg cache built on a 65nm process and no VT. It's not an Athlon II, just an Athlon 64 x2, I believe.On Cinebench R10:
Single Thread: E-350=1174, TK-42=1340 :-: 1174/1340 = 87.6%
2 Threads: E-350=2251, TK-42=2373 :-: 2251/2373 = 94.4%
(1174+2251=3425)/(1340+2373=3713) = 92.2%
Scaling seems better on the new chips than on the older ones.
At idle with the screen off the laptop pulls about 18 watts. In Cinebench on a single thread it pulls about 30 watts, with 2 threads it pulls about 33 wats. Opening the screen to run the LCD at full brightness adds about 9 watts at any time.
I ran these tests with a Kill-A-Watt meter. It's not quite an exact comparison, but is pretty close. But to see that they kept performance close, added graphics, and still managed to shave 10% off TDP it's pretty dang impressive.
lovansoft - Tuesday, February 1, 2011 - link
Actually, I just let my laptop sit idle for a while. Now, idle power usage dips down to 11 watts with the lid closed and generally stays switching between 11-13 watts. Hmmm, the power usage on these new chips aren't quite as I would expect unless it's a platform thing. This article shows that the new chips pull 9W at full load under Cinebench. My testing shows I ramp from about 12W up to 33W which is a 21W increase by taxing the TK-42, right inline with the 20W spec giving my rounding of numbers. All other parts being equal and I only ramp up 9W instead of 21W then my peak should be about 12W less, or about 21W total instead of 33W total. That would be a significant gain. Interesting that this article has the new platform at 32.2W with the same workload. That's about 50% higher than my rough estimations. Is it because it's a desktop board and not a laptop design?sebanab - Wednesday, February 2, 2011 - link
Thanks for the performance comparison. I really helps putting the Zacate into perspective.On the power consumption comparison:
Ofcourse the desktop board system will consume more than a laptop with same specs.
Your laptop consumes less because of different PSU , less USBs , less components in general (PCIex) and so on.
So in order to make a correct comparison , wait for a HP DM1z review for example.
lovansoft - Wednesday, February 2, 2011 - link
I figured as much, but 50% seems high. But then again, it really is only a few watts... Lower efficiency PSU, a couple more chips to provide some extra ports. A couple of watts here and there do add up I suppose. And to think it provides more than 90% of my current performance into only 2/3 the power. I'd think if they can up it to 2Ghz it'd be about the same as mine without tapping the power too much. From the speculation I've read, it makes it seem that the revision coming in a year should grow the performance by quite a bit without really increasing power, about what you'd expect from a die shrink.You know, with this architecture, it'd be nice if a board could be made that would have multiples of these chips for server use. From my experience in SMB, I rarely find servers being CPU bound. Usually if they are, then there is some runaway process that needs to be tamed. Maybe this current generation isn't quite fast enough, but with a process shrink and some speed adjustments, getting a few of these on a board would make a very low energy server. But it'd only be feasible if there were something they could use that built in graphics portion for. Otherwise it'd be a waste.
Oh, and Cinnebench R10 on an AthlonXP 3000+ (2Ghz) = 1438.
Single Thread: E-350=1174, XP3000=1438 :-: 1174/1438 = 81.6%
1.6/2=80% Seems to be about the same IPC as the AthlonXP line.
I don't have power numbers for it, though.
torkemada - Wednesday, February 2, 2011 - link
Is E350IA-E45 board HDMI 1.4? Info on the net is confusing.Hrel - Thursday, February 3, 2011 - link
that VIA chip is pretty impressive.