Low-Power Gaming: AMD's E-350 Vs. Intel Atom D525 And Ion 2

Page 3 - 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.

11796pcs

Distinguished
Jan 3, 2011
608
0
18,990
Perosnally I think a better RTS game to test would have been Sompany of Heroes- it's a perfect RTS game for LAN parties and can run on pretty low systems. Overall though I like this review and was pleasantly suprised to see some of the FPS reaching for 40 in some cases.
 

yyrkoon

Distinguished
May 8, 2006
387
0
18,780
[citation][nom]Cleeve[/nom]I think a better sum-up is to say you totally missed the point.If you actually read the first couple paragraphs you'd understand the purpose of the review. Hint: it's not to measure power consumption.[/citation]

That is the whole point of even bothering with one of these systems.


Unless you purport that they are cost, or performance effective( because they really are not ).

 

yyrkoon

Distinguished
May 8, 2006
387
0
18,780
[citation][nom]frederico[/nom]Examples please, and when including them, I want 10 inch and 11.6 inch only, under 1.5 kg, and under 350 dollars. Thanks.[/citation]

http://configure.us.dell.com/dellstore/config.aspx?oc=bqcwrmb&cs=04&dgvcode=ss&c=US&l=EN&dgc=BA&cid=61474&lid=1544287

Intel® Core™2 Duo Processor ULV SU7300 (3M Cache, 1.30 GHz, 800 MHz FSB)
13.3 inch Wide Display HD (1366 x 768) with Anti-glare
6-cell 30WHr internal battery
Intel Pro/Wireless 5100 802.11b/g/n
Mobile Intel Graphics Media Accelerator 4500MHD
US 65W AC Adapter
Dell Wireless 365 Bluetooth Module
1.3MP Webcam with Digital Mic for Win 7

13.3" screen, but as for asking for 10-11" inch only I think is a bit ridiculous. As is I think asking for something that weights less than 3.3 pounds. If you cant handle carrying 5 lbs around in a laptop case. Maybe you need to get a Gym subscription, or just carry a smart phone with you.

Power wise, I could not say, I do not own one. and do keep in mind I found it after 2 minutes of searching after seeing your post. I would wager a guess at 40-50W peak load though.

whatever the case, you're never going to get exactly what you want. Ifwe could, well then I'd have a machine right now that only used 1- watts, able to play any game, at any setting. As well as transcode full length DVD movies in under 5 minutes. Sound a little ridiculous to you? It does to me.
 

rdawise

Distinguished
Apr 18, 2008
225
0
18,680
[citation][nom]yyrkoon[/nom]That is the whole point of even bothering with one of these systems. Unless you purport that they are cost, or performance effective( because they really are not ).[/citation]

So you read the title "Low power: Gaming" and are upset because they actually showed the gaming results?

The author's point is not power efficiency it's gaming. The author already conceded that these systems are not meant for this, so it would be only logically to assume that the power consumption is irrelevant.

It would be like arguing that they didn't show you temperature ratings in an article that displayed the relationship between speed an engine size (yes I know temp affects gas consumption but that would be outside of the range of the article. my point exactly).

Great article. Very informative. Though I would have loved to see World of Warcraft added to the benchmarks.
 

belardo

Splendid
Nov 23, 2008
3,540
2
22,795
Unreal Tournament what?

Unreal Tournament 1999?
Unreal Tournament 2994
Unreal Tournament 3? (2008)

We have no frame of reference here. Somehow, I don't think its the 1999 game which ran fine on TNT2 / Voodoo2 cards with 16mb of video ram.
 

yyrkoon

Distinguished
May 8, 2006
387
0
18,780
[citation][nom]rdawise[/nom]So you read the title "Low power: Gaming" and are upset because they actually showed the gaming results?The author's point is not power efficiency it's gaming. The author already conceded that these systems are not meant for this, so it would be only logically to assume that the power consumption is irrelevant. It would be like arguing that they didn't show you temperature ratings in an article that displayed the relationship between speed an engine size (yes I know temp affects gas consumption but that would be outside of the range of the article. my point exactly).Great article. Very informative. Though I would have loved to see World of Warcraft added to the benchmarks.[/citation]

No, I am not upset, because of the included gaming results. I am not upset at all. My point of contention is that these results are incomplete without providing power consumption results. That is all.

Well, I will concede, that yes, because these systems will use more power than a laptop, that will vastly outperform them in most every task . . . Yes, power consumption then becomes irrelevant. At the same time, the performance tests are very likely related to each individual systems power consumption. When you're picking nits ( which is pretty much what this article, or any like it does ), that is important. Then considering the possibility that the under-performing system could even be using more power. . . it becomes even more important.

 
[citation][nom]masteren[/nom]Source?[/citation]

Well there are specs. The PS3 uses a modified (by that I mean weaker) 7900GTX while the 360 has a custom built ATI R500 chip that is above the X1950 in terms of performance. And the 7900GTX was in no way better than a X1950. A 8800 was but not a 7900GTX.

I just want to know why battery life is not included. Honestly, trying to game on Atom is stupid since it wasn't made for that but which system will last longer?

Even if Brazos gives 76% better performance, if it only lasts 25% of the time of the Atom platform whats the point?
 

yyrkoon

Distinguished
May 8, 2006
387
0
18,780
[citation][nom]jimmysmitty[/nom]Well there are specs. The PS3 uses a modified (by that I mean weaker) 7900GTX while the 360 has a custom built ATI R500 chip that is above the X1950 in terms of performance. And the 7900GTX was in no way better than a X1950. A 8800 was but not a 7900GTX.I just want to know why battery life is not included. Honestly, trying to game on Atom is stupid since it wasn't made for that but which system will last longer?Even if Brazos gives 76% better performance, if it only lasts 25% of the time of the Atom platform whats the point?[/citation]

Jimmy, these are nettops, not netbooks. Atom classed desktop systems.
 

pelov

Distinguished
Jan 6, 2011
423
0
18,810
The brazos draws less power than that atom combo, and significantly less during idle.

If you want the watt-rating just google it...
 

silverblue

Distinguished
Jul 22, 2009
1,199
4
19,285
[citation][nom]Belardo[/nom]Unreal Tournament what?Unreal Tournament 1999?Unreal Tournament 2994Unreal Tournament 3? (2008)We have no frame of reference here. Somehow, I don't think its the 1999 game which ran fine on TNT2 / Voodoo2 cards with 16mb of video ram.[/citation]
Page 3 lists it as Unreal Tournament 3. If it was any of the earlier editions, you'd be talking a far higher framerate (aside of max framerate figures, UT doesn't perform any better than 2003 or 2004 in all fairness).

[citation][nom]jimmysmitty[/nom]Even if Brazos gives 76% better performance, if it only lasts 25% of the time of the Atom platform whats the point?[/citation]

As pelov has said, Brazos not only performs better but uses less power. It's also cheaper than Atom + Ion 2 as there's one less chip to take into account when designing motherboards. Now, if Brazos was 32nm instead of 40, it'd be even more one-sided; either it'd use even less power, or AMD would've clocked it higher
 

mschlenker

Distinguished
May 7, 2009
16
0
18,510
I own an EEEpc 1215N (D525/ION2) and have watched the NVIDIA drivers mature over this year. Many 1215N owners have reported problems with SC2 since day one but the latest Bata drivers (270.51) have reportedly fix that and many other playability issues. I’m curious to know which Drivers were used in this test.
 

yyrkoon

Distinguished
May 8, 2006
387
0
18,780
[citation][nom]pelov[/nom]The brazos draws less power than that atom combo, and significantly less during idle. If you want the watt-rating just google it...[/citation]

Unfortunately, power readings as such would be inaccurate. You can not just take some arbitrary figures from some other source, and say "this is how much power I will use while playing WoW" or whatever.

There are many factors when figuring for power consumption. One of the biggest factors often overlooked is the power supply. Someone, may have the exact same system setup, with only a different power supply, and there can be a huge power consumption difference between the two. Especially when comparing low power (electricity ) systems such as these. Even the actual voltage being transported throughout your home can make a difference in power consumption ( if there is a difference ).

So yes, it may be impossible to exactly duplicate one persons observations in some, or all situations. But the more complete the test results we are given, the more complete picture we all get.

Otherwise, we end up with a difference similar to the difference between Astrology, and Astronomy. In case I need to point it out. One is science as we know it today. Where the other is not. e.g. factual vs fiction.
 

icebug

Distinguished
May 5, 2009
16
0
18,510
I will be buying a laptop here in the next few months and I love the portability of some of the E-350 systems. One game that I actually would be playing is LOTRO (lifetime subscriber since launch) so this makes me happy to see it included in the roundup. One thing that I see wasn't mentioned here in the review was whether or not the E-350 set-up was running the game in DX11 mode or not. I know that running LOTRO in DX11 GREATLY increases framerates from the same settings on DX10 but does that equal more frames on this system as well?
 

pelov

Distinguished
Jan 6, 2011
423
0
18,810
[citation][nom]yyrkoon[/nom]Unfortunately, power readings as such would be inaccurate. You can not just take some arbitrary figures from some other source, and say "this is how much power I will use while playing WoW" or whatever.There are many factors when figuring for power consumption. One of the biggest factors often overlooked is the power supply. Someone, may have the exact same system setup, with only a different power supply, and there can be a huge power consumption difference between the two. Especially when comparing low power (electricity ) systems such as these. Even the actual voltage being transported throughout your home can make a difference in power consumption ( if there is a difference ).So yes, it may be impossible to exactly duplicate one persons observations in some, or all situations. But the more complete the test results we are given, the more complete picture we all get. Otherwise, we end up with a difference similar to the difference between Astrology, and Astronomy. In case I need to point it out. One is science as we know it today. Where the other is not. e.g. factual vs fiction.[/citation]

It does consume less, and it's not some arbitrary number. The reviews that I've read are referring to the mini-ITX format, where you get the mobo+APU+cooler (e-350) and compare it to the mobo+CPU+gpu+cooler (atom+ion) using the same system specs and the same power supply. These apu's aren't available only in netbook/nettop/laptop forms, in fact you can build your own systems and HTPCs around them. They're benchmarked the same way every other piece of hardware is. To imply that somehow it's wrong here and right everywhere else is nuts, and to imply that it's wrong everywhere is just as crazy. The reviews I've read (probably 4-5) all state the same in terms of power use: it consumes less power.
 

seriesshorty

Distinguished
Mar 31, 2011
4
0
18,510
This article couldn't have come at a better time. I was very close to purchasing the Atom/Ion based system 2 weeks ago, then I read about the E350's in CPU magazine and started leaning that direction. I just got back from Microcenter with an MSI E350 motherboard in hand and am settling down to build an HTPC with it. I feel much better about my choice after reading this article.
 

Arbie

Distinguished
Oct 8, 2007
208
65
18,760
Thanks for the review! It's just the sort of thing I'd like to know. Netbook size is very good for me since I need a usable keyboard and the longer battery life over a tablet, but want something much smaller than a laptop. And the netbook doubles as entertainment for the kids and sometimes the adults too. So while we don't buy it for gaming, we'll certainly favor a machine that can do so.

I just wish you had included Crysis / Warhead. That to me is the defining PC game to date, and it was wonderfully scalable. You might have found it playable at minimum settings, which would be something to know.
 

BSMonitor

Distinguished
Nov 19, 2007
167
0
18,680
Love charts like the conclusion. "OMG look at that big blue bar, 300% better WOWOWOW!!!!@@!" .. Great, the blue bars from the previous 20 slides show ALL of the games being unplayable or lots of lag. So the Atom is 3 times more not playable than the E-350(also not playable).
 

ra3tonite

Distinguished
Feb 2, 2010
34
2
18,535
[citation][nom]warjunkieltu[/nom]Why are you guys not testing some good quality freeware first person shooters, that could run smoothly on those low end PCs? For example:Urban Terror (my favorite one), Open Arena, Alien Arena? Or at least Quake live? Cause it's ridiculous to test all those latest shooters on such low end machines...[/citation]

did somebody say Quake Live?
 

masteren

Distinguished
Mar 27, 2011
5
0
18,510
[citation][nom]masterofevil22[/nom]I don't know the source, but this sounds about right. PS3's nvidia chip has 24 fixed shaders (8 pixel, 16 vertex i think), Xbox 360 Xenos has 48 unified and the Wii (so far as I know) has a grand total of ZERO. Even the Xbox 1 had 4 (Gamecube and Ps2 were zero also).[/citation]

This is not a 1:1 comparison. You are practically saying: You have a PC/Console/Thing with 4 cores but only 4 MB L3 Cache. I have a PC/Console/Thing with 2 cores, but with 6(!!!) MB of L3 Cache. This means that my PC is better than yours, because you only measure performance on one Parameter.

See? Wrong.
 
Status
Not open for further replies.