Archive Home arrow Reviews: arrow Memory arrow G.Skill ECO 4GB DDR3 Memory Kit F310666CL7D
G.Skill ECO 4GB DDR3 Memory Kit F310666CL7D E-mail
Reviews - Featured Reviews: Memory
Written by Servando Silva   
Monday, 10 May 2010
Table of Contents: Page Index
G.Skill ECO 4GB DDR3 Memory Kit F310666CL7D
G.Skill ECO F3 10666CL7D Specifications
Closer Look: G.Skill ECO DDR3
G.Skill ECO SPD
RAM Testing Methodology
Benchmark Tests Results
Application Performance Results
Overclocking and Power Consumption
Final Thoughts and Conclusion

Benchmark Tests Results

Benchmark Reviews runs each tests at least 3 times to avoid special situations where scores are over/under exposed. Passmark Performance Memory Test does some bench to the memory and reports a test score. I´ve chose Memory Mark along with uncached read and write speed since they´re the most important benchmarks for memory testing on this suite.

GSkill_ECO_Passmark_Memory.png

Increasing frequency up to 1600MHz shows a little advantage on the final results. In this case, 1600MHz 6-7-6-21 is the best configuration for maximum performance, and that will probably be the best configuration for all the tests done today since it combines decent high frequency with relatively low timings.

EVEREST Ultimate Edition offers three simple memory bandwidth tests that focus on the basics; Read, Write, and Copy. In order to avoid concurrent threads competing over system memory bandwidth, the Memory benchmarks utilize only one processor core and one thread.

GSkill_ECO_Lavalys_Everest.png

Again, Lavalys Everest results increase with higher frequencies and tighter latencies. There is almost 12% performance increase with 1600MHz in each test. Considering how easy is to achieve this configuration (and probably it can be done with lower voltage in the right motherboard) it is definitively the best way to use your G.Skill ECOs.

Sandra is based on STREAM, a popular memory bandwidth benchmark that has been used on personal computers to super computers. It measures sustained memory bandwidth not burst or peak. Therefore, the results may be lower than those of other benchmarks. STREAM 2.0 uses static data (about 12M) - Sandra uses dynamic data (around 40-60% of physical system RAM). This means that on computers with fast memory Sandra may yield lower results than STREAM. It's not feasible to make Sandra use static RAM - since Sandra is much more than a benchmark, thus it would needlessly use memory.

A major difference is that Sandra's algorithm is multi-threaded on SMP/SMT systems. This works by splitting the arrays and letting each thread work on its own bit. Sandra creates a thread for each CPU in the system and assigns each thread to an individual CPU. Another difference is the aggressive use of scheduling/overlapping of instructions in order to maximize memory throughput even on "slower" processors. The loops should always be memory bound rather than CPU bound on all modern processors.

GSkill_ECO_SiSoftSandra.png

Sandra shows up to 40% performance increase when overclocking up to 1600MHz. The difference gained by tightening memory latencies makes little to no change in this benchmark. From all these Benchmark results, the conclusion is simple: 1600MHz 6-7-6-21 1T will give you the best performance sacrificing 150 mili-volts. While I wasn´t able to test voltages between 1.35v and 1.5v, the G.Skill F3 10666CL7D were very stable at this setup, so they might be able to operate at 1600MHz and tight latencies with lower voltage... who knows?



 

Comments 

 
# Excellent Review!Justin MacQueen 2010-05-09 23:02
Hi Servando, an excellent review! I have been running two of these kits (8GB total) in my rig for some time, on an ASRock P55M-Pro mATX board with an i7 860. Right from the get-go they booted at correct timings and correct voltage. I'm very happy with them myself. Your overclocking results are interesting, I'm going to go and try your 1600MHz settings right now and see whether two kits are stable at these timings or not. Looking forward to seeing how you get on with an i7! Also for reference, my sticks use the Elpida BBSE you mentioned. Thanks again!
Report Comment
 
 
# RE: Excellent Review!Servando Silva 2010-05-10 17:44
Thanks Justin,
Since you have some Elpida BBSE you should go and try rising up the tRCD as I did. I really loved the 1600MHz 6-7-6 configuration. Just to let you know, I´ve got a i5 750 paired with the P55 ASUS Maximus III Formula, and I´ll update my results as soon as I´ve a time (probably on the forum thread, but I´ll link it here). Right now, they´re working at 1.3v with stock settings without problems, but I wasn´t able to boot at 1.25v.
Report Comment
 
 
# Commendations again on a review well done!K Gregory 2010-05-10 12:10
Sharp review again here at Benchmark Reviews. I came across a review of the Kingston Hyper X Low Voltage product and I was very impressed. I think the G.Skill Eco has the better pricing vs the Kingston Hyper X 1333 kit($14 between the products)and slightly better timings. -Great that you added those superior timings in your testing that were achievable at such a low voltage!
Report Comment
 
 
# RE: Commendations again on a review well done!Servando Silva 2010-05-10 17:41
Hi Gregory!
I´ve seen the Kingston Lo-Vo Edition and I think this kit is better. G.Skill memory will probably overclock better and they cost less as you´ve said. Kingston in the other hand, has a kit that can work with 1.25v but at expense of higher latencies.
Report Comment
 
 
# Thank you for this great review.B Hin 2010-05-13 11:36
I have been using the same memory for a while now. Using your results, I have tried to find the lowest voltage at which I could run it.
At 1.28V with the settings 7-8-7-21 1t at 1333MHz the memory passes several hours of stress testing. (I didn't test any longer, it didn't fail)

I didn't manage to get it stable at a lower voltage, however the system did boot on 1.26V.
Tested on an ASUS P7H55D-M EVO - motherboard, with a core i3 530 running at default clock.
Report Comment
 
 
# RE: Thank you for this great review.Servando Silva 2010-05-13 13:43
I did the same low voltage tests on the Maximus III Formula. This kit in particular works at 1.3v (set on BIOS) or 1.27v (measured with multimeter).
As you said, anything lower than that won´t be stable or barely boot windows. BTW, I used 7-7-7 timings.
Report Comment
 
 
# Additional findingsB Hin 2010-05-15 14:51
Aside from the normally clocked settings, I also have a profile at which the i3 530 is overclocked to 4400 MHz. The idea being that one push of the powerbutton turns a very economical computer into something a bit more powerfull. (Although at the current version of the asus software for this motherboard the profile switching feature doesn't seem to work yet, unfortunately.)
Report Comment
 
 
# Additional findingsB Hin 2010-05-15 14:52
For the overclocked profile I decided to find the tighest settings at the highest memory frequency possible. Which is 1600 for this processor. (with the help of the motherboard, normally it is 1333)
I turned up the frequency to 1600, and the voltage to 1.56V to find the tightest settings. I could only get it to work at 7-8-7-21 1T Which at first disappointed me a little.
I then slowly decreased the voltage to find the minimum at this setting. Now this part really surprised me: the memory remained stable (even after several hours of stress testing) at only 1.34V
What is really amazing about that is that it is still below the voltage setting of the XMP for 1333MHz.
So in recap,
min (core i3 @2.9MHz): 1333 MHz @ 7-8-7-21 1t -->1.28V
max (core i3 @4.4Mhz): 1600 MHZ @ 7-8-7-21 1t --> 1.34V
Report Comment
 
 
# RE: Additional findingsServando Silva 2010-05-15 15:35
Well, that is really weird. For 1.28v, have you tried 7-7-7-21 1T?
And what´s the best you can do (MHz) at 6-7-6-21 1T? I can´t think your memory does the same at 1.35v than at 1.55v. There should be some kind of limitation. It could be your MB (not probable), your memory (it might be), or a bad CPU for Mem scaling.
I still haven´t tried max freq scaling with this motherboard since work is drowning me but I bet it will do better than the Mini-ITX I used for the review (it better do it).
Take care.
Report Comment
 
 
# reB Hin 2010-05-17 10:35
Thanks for pointing out that something must be limiting the voltage scaling, I will look deeper into it when I have the time. I did indeed find that the memory remains stable at 7-7-7-21 1T @ 1.28V.
Report Comment
 
 
# RE: G.Skill ECO 4GB DDR3 Memory Kit F310666CL7DServando Silva 2010-05-18 20:07
I just want to add. I did tests on the Maximus III Formula, and other than discovering this kit works perfectly at 1.3v, I couldn´t achieve better OC results, neither better timings. It seems this is the MHz wall of this kit. The 1600MHz kit, should do a little bit more.
I also discovered 1600MHz 6-7-6-21 1T needs only 1.4v (multimeter) and not 1.5v. But the Mini-ITX used didn´t let me test that. This is still a great kit.
Take Care.
Report Comment
 
 
# XMP on H55H-IDean 2010-08-27 22:28
Based on your review I bought the H55H-I MB and gskill eco. Only problem I have is I cant get the MB down to 1.35V. Can you tell me where th XMP setting is in BIOS. Many thanks.
Report Comment
 
 
# RE: XMP on H55H-IOlin Coles 2010-08-27 22:31
Dean, this setting is usually in the memory configuration of the BIOS, under advanced settings. Your manual should be able to help.
Report Comment
 
 
# RE: G.Skill ECO 4GB DDR3 Memory Kit F310666CL7DRuzveh 2010-08-31 19:26
Hey guys just a simple suggestion in each of your reviews posted on the website. Whenever you mention the price on the conclusion page please mentioned date on which the test was posted. Because i dont knw when the test was done for which the price is $xxx. Hence i cant not judge properly whether its a good value buy or not worth.
Report Comment
 

Comments have been disabled by the administrator.

Search Benchmark Reviews
QNAP Network Storage Servers

Follow Benchmark Reviews on FacebookReceive Tweets from Benchmark Reviews on Twitter