Sign in with
Sign up | Sign in
Your question
Closed

Does High-Speed DDR3 Help AMD's FX? Four 8 GB Kits, Reviewed

Tags:
Last response: in Reviews comments
Share
June 1, 2012 5:50:36 AM

Interesting!. Still would wait for Pile driver. But since i do have a Kingston 8GB kit and a Sabertooth 990FX... may just have to try it one day
Score
2
June 1, 2012 6:11:23 AM

Great article this proves a lot of rumors!

People can finally stop saying how unfair it is to test Amd with slower ram when it supports higher speed ram even though higher speed ram cost more money. It would actually be unfair to Intel its not their fault Amd needs higher speeds to compete(which it doesn't) with Intel's 1333 ram controller.

Ram speeds do how ever make a big difference on APU's.
Score
16
Related resources
June 1, 2012 6:12:03 AM

would love to see the same review for Ivy Bridge processors with and without iGPU HD4000.

and do include more tests in the review.
Score
11
June 1, 2012 6:21:49 AM

More of the cheap stuff is my mantra.
Score
7
June 1, 2012 6:27:05 AM

sarinaideMore of the cheap stuff is my mantra.
Yeh yeh, but I really can't complain about paying $70 for 8GB of DDR3-2133, especially when it overclocks like that.
Score
8
June 1, 2012 8:18:30 AM

AMD has enjoyed the performance benefits of an integrated memory controller for more than twice as long as Intel. And yet, it seems that Intel sets today's standard for acceptability.

Replace the last word with "performance" crash.

Good article which gets answers a couple of questions people have been asking - well done !!

:) 
Score
3
June 1, 2012 8:32:01 AM

Thomas thanks for the review.

First note:
*KHX2400C11D3K4/8GX - is a Quad Channel kit specifically designed for the X79 ; link http://www.kingston.com/us/company/press?pagename=n1111...
*WQ213UB4G vs WQ213UX16G (kit) - is ONE stick of RAM (not a matched kit) and its 'kit version' is Quad Channel kit specifically designed for the X79 ; link http://www.supertalent.com/datasheets/WQ213UX16G.pdf

After that it makes sense that those (2) "kits" (in one case 4 individual sticks) failed @ Rated, and I wasn't interested in researching the others. I've seen other more in-depth testing on the AMD FX-8150 indicating that overall there's a slight advantage for the DDR3-1866 kits on a most of the AMD FX processors. Though @ 4AM 'to me' I'll edit this post after a few cups of coffee.

The 'ideal' AMD kits have JEDEC 'SPD' @ Rated 'cloned' (XMP) encoding which is optimized for AMD and sometimes slightly different so you're not 'stuck' as you put it @ DDR3-1333. Those are the better KITS to test on the AMD FX lines.
Score
3
June 1, 2012 8:35:39 AM

Ram sppeds on a Bulldozer CPU means little and this just proves that so it doesn't matter if you are using DDR3 1600 ram or buying high performance RAM that can overclock to DDR3 2800 your not really seeing hardly any improvement with Bulldozer by doing so.
Score
7
June 1, 2012 8:49:15 AM

Good article, very informative!
Thanks :) 
Score
3
Anonymous
June 1, 2012 9:06:54 AM

RAM speed also has a HUGE impact in controlled cache environments:
http://thessdreview.com/our-reviews/romex-fancycache-re...

Even the best SSDs can barely muster 100,000 IOPs, yet even slow RAM can easily exceed 700,000 IOPs.
Bandwidth, throughput, and latency are equally insane. RAM puts SSDs to shame exponentially more than SSDs put hard drives to shame. Contrary to what many "enthusiasts" believe, RAM timings are virtually irrelevant. Frequency is by far, much more important.
Score
0
June 1, 2012 9:09:49 AM

jaquithThomas thanks for the review.First note:*KHX2400C11D3K4/8GX - is a Quad Channel kit specifically designed for the X79 ; link http://www.kingston.com/us/company [...] glish_emea*WQ213UB4G vs WQ213UX16G (kit) - is ONE stick of RAM (not a matched kit) and its 'kit version' is Quad Channel kit specifically designed for the X79 ; link http://www.supertalent.com/datasheets/WQ213UX16G.pdfAft... that it makes sense that those (2) "kits" (in one case 4 individual sticks) failed @ Rated, and I wasn't interested in researching the others. I've seen other more in-depth testing on the AMD FX-8150 indicating that overall there's a slight advantage for the DDR3-1866 kits on a most of the AMD FX processors. Though @ 4AM 'to me' I'll edit this post after a few cups of coffee.The 'ideal' AMD kits have JEDEC 'SPD' @ Rated 'cloned' (XMP) encoding which is optimized for AMD and sometimes slightly different so you're not 'stuck' as you put it @ DDR3-1333. Those are the better KITS to test on the AMD FX lines.
Read page 2. They're all Intel-optimized kits because that's the only thing any of these companies have. If you read something about an AMD-optimized kit being for sale somewhere, it's probably either a niche manufacturer or simply a marketing lie.
Score
9
June 1, 2012 9:10:19 AM

If you're going to spend extra money on memory, why not use the extra money for a real CPU? Meaning, from Intel, AMD doesn't make anything remotely competitive, and if you're getting performance parts to try to make it perform, you're probably better off starting with a processor that performs better in the first (second, and third) place.

It's like putting a diamond in an aluminum ring. It makes no sense.
Score
-6
a b À AMD
June 1, 2012 9:12:29 AM

fx needs ddr3 2133 kits to reach 20~ gb/s on sandra bench with while a sandy bridge pentium can reach the same bw @ddr3 1333.
http://media.bestofmicro.com/G/Y/327202/original/sandra...
amd apus are different story. those perform better in games with faster memory.
only posted in case some wonder how much intel gets out of rams.
Score
7
June 1, 2012 9:27:59 AM

noob2222great article, 2 games and 2 programs that don't respont to memory speed must represent all games ever made. This article proves one thing, not every program responts to memory speed.http://vr-zone.com/articles/amd-fx [...] 13704.htmlhttp://www.madshrimps.be/articles/ [...] z1wX1gAnfFIll match your 4 cherry picked benchmarks against 2 websites that state otherwise.Throw civ V into the mix and your way off.
Cherry picked for what, to prove that faster memory helps? Well guess what, it didn't help much. Or did you mean cherry picked for this article? Are you making blind accusations without even looking to see which benchmarks were used in the last several memory articles?

Please, at least try to be honest. You already knew the Madshrimps article was explicitly dishonest because it used CAS 8 timings at all speeds (even 2133). Real-world memory supports tighter timings at lower frequencies, and requires looser timings at higher speeds. Madshrimps gimped the low-speed tests and boosted the high speed configuration intentionally.
Score
10
June 1, 2012 9:32:46 AM

take a small enough sample to make sure the results don't prove a thing. why not use some programs that do respond to memory speed?


Score
6
June 1, 2012 9:44:21 AM

noob2222take a small enough sample to make sure the results don't prove a thing. why not use some programs that do respond to memory speed?
The test here was set up long ago to use 50% GPU-limited and 50% platform-limited games. Replace GPU with CPU for the other tests, and the hope was to find 50% of the benchmarks with noticeable performance gains. Of course that's based on a real-world scenario, where both GPU and platform-limited games coexist.

Notice in the chart you showed, the difference between 2133 and 1866 is rather small. The baseline in this test was DDR3-1600 CAS 9 because that's the slowest memory a performance builder would use. In other words, the 1333 and 1066 results are irrelevant.

The benefit you're seeing in some articles is reduced real-time latency, which I'm fairly certain is discussed in this article. DDR3-1333 CAS 7 has the same real-time latency as DDR3-2666 CAS 14, since the cycle time is inverse of frequency.

If we assume for only a moment that DDR3-1600 is barely fast enough to fill the bandwidth needs, most of the difference over 1600 actually measures how long the CPU is waiting for instructions. Now I wouldn't actually assume 1600 is enough, I only assume that 1600 is the minimum speed that goes into most performance builds. After all, who would buy 1333 to save money in a performance machine when 1600 is so cheap?
Score
14
June 1, 2012 9:52:33 AM

CrashmanThe test here was set up long ago to use 50% GPU-limited and 50% platform-limited games. Replace GPU with CPU for the other tests, and the hope was to find 50% of the benchmarks with noticeable performance gains.


the problem is both of those games, dirt 3 and metro 2033 are gpu limited. http://www.techspot.com/review/403-dirt-3-performance/p...
Score
-6
June 1, 2012 9:55:30 AM

noob2222the problem is both of those games, dirt 3 and metro 2033 are gpu limited. http://www.techspot.com/review/403 [...] page7.html
GPU limited at max quality and 1920. This test was at medium resolutions and details (the "high" setting in these games is actually mid-high). DiRT 3 in particular is supposed to be GPU-bottlenecked only at its highest settings.
Score
10
June 1, 2012 10:05:24 AM

If you have a copy of civ V, try it, you will laugh at the results, 1333 to 2133 is nearly 25% faster on the full render. There are benefits to faster memory, limiting the benchmarks limits the results.
Score
0
June 1, 2012 10:12:43 AM

CrashmanRead page 2. They're all Intel-optimized kits because that's the only thing any of these companies have. If you read something about an AMD-optimized kit being for sale somewhere, it's probably either a niche manufacturer or simply a marketing lie.

Thomas -- I have seen (with my own eyeballs) the AMD so called 'optimized' kits with cloned e.g. DDR3-1866 with BOTH XMP and 'SPD' JEDEC (Duplicate) encoding.

No LIE -- Google for yourself or look at one of my (1000's) of RAM posts.

I of all folks here have seen and know plenty especially when it comes to RAM -- name another 'here @ TH' that knows more.

I can only 'wish' one size fits all when it comes to RAM -- it clearly does not! Adjust the OC, CAS Timings and Voltages i.e. take some time and I'm confident unless those (2) I looked a in some way 'damaged' (possibility) can & will work -- try raising the CAS if Voltage fails -- play around in the BIOS.

Frankly, I NEVER advise anyone, especially with >DDR3-1333 (even then I still do), to purchase anything BUT a fully matched set. G.SKILL in particular doesn't want to RMA (1) stick from e.g. an 8xDIMM fully matched set especially as either the Frequency <or> density is >4GB/stick. Heck the (stable) CAS per Frequency often doesn't work when you put (2) two Xerox (same PN) sets together.

I am not looking to debate you, I linked B&W the info on the kits.
Score
0
June 1, 2012 10:27:03 AM

jaquith said:
Thomas -- I have seen (with my own eyeballs) the AMD so called 'optimized' kits with cloned e.g. DDR3-1866 with BOTH XMP and 'SPD' JEDEC (Duplicate) encoding.

No LIE -- Google for yourself or look at one of my (1000's) of RAM posts.

I of all folks here have seen and know plenty especially when it comes to RAM -- name another 'here @ TH' that knows more.

I can only 'wish' one size fits all when it comes to RAM -- it clearly does not! Adjust the OC, CAS Timings and Voltages i.e. take some time and I'm confident unless those (2) I looked a in some way 'damaged' (possibility) can & will work -- try raising the CAS if Voltage fails -- play around in the BIOS.

I am not looking to debate you, I linked B&W the info on the kits.
No, you didn't link the info on "the" kits, because "the" kits are the nonexistant kits you speak of. You linked to "these" kits, which, like any other performance RAM, are Intel optimized.

You can't sucessfully SPD an XMP value for 1.65V, because SPD has no voltage info. If you did, it still wouldn't be JEDEC approved.

Maybe you should read Page 1: All major manufacturers got invited, all were told this was an AMD article, and all but four of these companies said something like "Sorry, we don't have any memory that's a good match to the article" (that response is implied for companies that refused to respond). That should have put an end to your argument, yet here we are having this discussion.

I'm not trying to be crass, but the stuff I'm currently going over with you is stuff I went over on my own before testing even began.
Score
4
June 1, 2012 11:15:24 AM

I don't know what you're talking about <or> you simply didn't read what I said. Existent or not the links are to the RAM you have placed in your article, and IF you don't like what's there take it up with the Vendors. I assume you told the folks for the 'free' RAM that you were testing on the AMD platform and IMO there was a terrible miscommunication.

Before I was referring to kits with 'matching' JEDEC / XMP info like shown below (Example of a DDR3-1600 kit with duplicate profiles for DDR3-1600):


I don't need a headache this AM...it's nothing I care too much about, and I was researching 'why' the (2) 'sets' failed @ Rated and at least 'I' get it...
Score
1
June 1, 2012 11:53:17 AM

jaquith said:
I don't know what you're talking about <or> you simply didn't read what I said. Existent or not the links are to the RAM you have placed in your article, and IF you don't like what's there take it up with the Vendors. I assume you told the folks for the 'free' RAM that you were testing on the AMD platform and IMO there was a terrible miscommunication.

Before I was referring to kits with 'matching' JEDEC / XMP info like shown below (Example of a DDR3-1600 kit with duplicate profiles for DDR3-1600):
http://aphnetworks.com/review/g_skill_sniper_f3_12800cl9d_8gbsr2_2x4gb/cpuz.png

I don't need a headache this AM...it's nothing I care too much about, and I was researching 'why' the (2) 'sets' failed @ Rated and at least 'I' get it...
You're showing me a JEDEC DDR3-1600 CAS9? Here I thought we were talking about performance RAM.

Since we're talking about two different things, there's a good chance that we're not actually in disagreement.

As for the 1.28V thing, well, I actually left the JEDEC updates for low-voltage RAM out of this discussion of non-low-voltage RAM. Since I'm a stickler for details, I apologize for skipping over that detail.
Score
5
June 1, 2012 12:28:52 PM

CrashmanYou're showing me a JEDEC DDR3-1600 CAS9? Here I thought we were talking about performance RAM.Since we're talking about two different things, there's a good chance that we're not actually in disagreement.As for the 1.28V thing, well, I actually left the JEDEC updates for low-voltage RAM out of this discussion of non-low-voltage RAM.

Find posts of mine @ TH where I'm recommending >DDR3-1600 CAS 9 RAM; it'll be a long l-o-n-g search. Laser-focused on AMD FX CPU lines 9/10 times I'll recommend DDR3-1600 CAS 9 and in a few very rare cases where the OP 'might' get a tad extra performance (based on use/request - after I've lectured the $ per performance) then maybe DDR3-1866 i.e. native to their CPU.

To clarify, I had (2) separate ideas and statements:
1. Possible reasons the (2) kits failed
2. AMD 'optimized' RAM (like the picture above)

Real world, run RAM Disk and there you'll find some gains from the increased memory frequency relative to the CPU in question and it's rated 'Max Memory Bandwidth' per spec.

My observation, again, is the (2) sets that failed and the probable reasons for failure (i.e. compatibility). Mainly their Frequency to CAS isn't ideal for the AMD CPU that you tested -- that's it for this wild goose chasing argument.

And yep I like both DDR3L & DDR3U IF your CPU can run them! :) 

In the 'Big Picture' I really don't care. AMD has always had a mediocre and sensitive IMC, and NEVER do I recommend DDR3-2400 to anyone -- I don't even care for those sets -- too much a PITA and too damn many BSOD's for my appetite regardless of Intel (SB, SB-E or IB) or for sure AMD FX. Yeah, I can get them to work but since I use RAM Disk I don't want the risk of e.g. outputting a couple hours of SQL 'garbage' or other failure. Gaming there are multitudes of benchmarks already out there.
Score
0
June 1, 2012 1:37:02 PM

Honestly this is the first I have ever heard of faster RAM benefiting Bulldozer.

I always knew/thought it helped AMD APU's tremendously.
Score
3
June 1, 2012 1:51:52 PM

noob2222take a small enough sample to make sure the results don't prove a thing. why not use some programs that do respond to memory speed?


Throwing a benchmark with DDR3-1066 on an i7 K processor is like buying a 680 and powering it with a Diablotek 350w PSU.
Seeing that most people with an i7 K processor is most likely looking already at DDR3-1600, there really is NO significant difference. Unless you think 1.16fps is worth the cost of buying DDR3-2133 ram.
Score
2
June 1, 2012 2:18:51 PM

Good to know. I just bought an fx8150, and am running DDR3 1600. Definitely a word of consolation that I'm not losing out by not getting higher-end ram.
Score
3
June 1, 2012 2:39:56 PM

I would have liked to see this memory performance comparison run on an APU instead of using a discrete GPU. In particular, a recent Intel vs AMD APU article hinted that the AMD APU could do better with faster RAM. Here's some much faster RAM - let's see what it does for APU performance, both for games, and again for apps.
Score
2
June 1, 2012 3:05:24 PM

yeah faster memory for apus with low-mid graphics could be better... but still it can only do so much.
Score
2
June 1, 2012 3:10:36 PM

i wonder if any of those kits would allow a higher cpu overclock than the others?
Score
1
June 1, 2012 3:36:12 PM

Love my Gskill Ripjaws :)  good RAM
Score
2
June 1, 2012 3:41:38 PM

I am sure we could argue about which is better, intels memory controller/ performance boost, vs AMD's mem controller/performance boost. But can we just get along and have a positive discussion?
Score
0
June 1, 2012 4:07:42 PM

I don't think the performance increase justifies the cost increase to be honest. Unless we have a fairly comprehensive list of applications that benefit from higher RAM speeds, I think it's better to stick with around 1833.
Score
2
June 1, 2012 4:08:59 PM

60% more cost for 6% performance gain? On one application tested? I'll pass. I don't care if it's only $25 more. That's 5 to 6 more beers I can get. Priorities, man. Priorities.
Score
0
June 1, 2012 4:48:32 PM

Wouldn't it have been better, since this was about improving Bulldozer CPUs, to have included some game benchmarks that were CPU-dependent as well? Because I thought Metro and Dirt3 (despite being a 'physics-based' game) were relatively lenient on the CPU.

It's definitely good to have those two though, to show whether a change would occur in something less affected by CPU performance. Overall I thought it was a good article.
Score
0
June 1, 2012 5:23:37 PM

sad thing is.. it does not matter... thus I will keep my oc'd 1100t @4Ghz, 3200HT. 1600ddr3 88824 (Btw when i benched the 8150fx in my system it was slower ( but used less power at idle 10 watts) and i could not oc'd it to 4.9-5.0Ghz which would have made the 8150 faster... so I wish for the day that i can replace my system with an intel 3770...

Score
2
June 1, 2012 7:02:33 PM

1. Your charts are useless since you are comparing RAM of identical performance from different menufacturers (which, lo and behold, gets identical performance in testing!) rather then comparing RAM of different performance levels (aka, each chart should contain DDR3-1600, 1866, 2133, 2400 compared for a singular test).

2. Does High-Speed DDR3 Help AMD's FX? No, nothing can help the abortion that is BD
Score
1
June 1, 2012 7:51:40 PM

Interesting to say the least. I guess for AMD it would just be better to get the best priced RAM from a reputable manufacturer.
Score
1
June 1, 2012 7:52:22 PM

CrashmanPlease, at least try to be honest. You already knew the Madshrimps article was explicitly dishonest because it used CAS 8 timings at all speeds (even 2133). Real-world memory supports tighter timings at lower frequencies, and requires looser timings at higher speeds. Madshrimps gimped the low-speed tests and boosted the high speed configuration intentionally.


I found a real gem in their article about going from 2.2GHz NB up to 2.6 and 3.2 and the approx. 2.5% performance boost an entire 1GHz of NB provides:

Quote:
Nice scaling again with higher NorthBridge speeds.


Sure, it might make more of a difference in actual games, but is the noticably higher power consumption worth 5%? You'd do far better just increasing the multiplier and going from stock to 4GHz.

In addition, I should add that the article goes all NVIDIA with some of the bar graphs which can indeed emphasise difference far more than it actually should.

Bottom line - Bulldozer is the first incarnation of a very different architecture, and the one surefire way of improving performance with it is to increase clock speed. Future incarnations should be more efficient but for now, this is what we have. Assuming you take the limit off the CPU, it won't perform badly, but drop details or add in a second GPU and it just cannot match even Intel's mid-range models. AMD knew it wouldn't meet expectations... hence the focus on "unlocked" and the liquid coolers. Piledriver won't close the gap to Intel massively, but you can be sure that it'll be easily quicker and more frugal than Bulldozer.

And whatever happened to that B3 stepping, anyway?
Score
-1
June 1, 2012 8:42:01 PM

I would gladly pay $25 for a 6% speed increase on my computer, but then again I wouldn't buy a bulldozer unless I was running linux (piledriver for win8 maybe?) and would instead invest that $25 into an Intel chip.

Pretty much every benchmark points out, that an I-3 at 3.1Ghz is fast enough for basically anything you can throw at it for 100% of consumer grade computers. All in all the AMD chips while they aren't the king of performance, do just fine for 95% of the computing world, and I see them as less of a failure than netburst. (1.4Ghz p4 slower than 800Mhz p3 anyone?)

I am enjoying the time of fast CPU's for $200, and will dread the day AMD stops competing against Intel on the high end, as we will quickly return to the > $1000EE cpu days.
Score
2
June 2, 2012 4:58:01 AM

If AMD doesn't improve the IPC performance of their CPUs, they'll be out of business. I really like and respect AMD. But, right now, I'll be buying Ivy Bridge. I hope AMD has a good CPU by 2015 or I'll be forced to buy Intel again. The same applies to GPUs, if their GPUs continue to get outperformed by Nvidia at the same price. I'll buy Nvidia. Now that I'm gaming at 1080p, I can't afford to buy lower performance for more money
Score
5
June 2, 2012 6:19:36 AM

I think the rather high cache latency in the CPU really negates the benefit of the faster RAM. That or the memory controller. Kinda disappointing. Shame on you AMD, shame. Oh well
Score
3
June 2, 2012 8:05:44 AM

Quote:
Pretty much every benchmark points out, that an I-3 at 3.1Ghz is fast enough for basically anything you can throw at it for 100% of consumer grade computers. All in all the AMD chips while they aren't the king of performance, do just fine for 95% of the computing world, and I see them as less of a failure than netburst. (1.4Ghz p4 slower than 800Mhz p3 anyone?)

lol 100%? Not everyone is only surfing/playing games. And even gamers... not everyone plays badly written games, which can only use one or two cores.

P4 was slower because of its extra-long pipelines compared at the same clock, not at half speed. lol
Score
0
June 2, 2012 1:22:10 PM

Fx cpu's overclock nicely :D 
Score
0
June 2, 2012 6:09:02 PM

i guess high speed ram will show a different in folding@home.
hope to see folding@home benchmark with high speed ram.
Score
2
June 3, 2012 2:17:01 PM

Would It be possible to see one of these kits working with a APU to see how the CPU-GPU combo benifits from fast ram, and 8GB Vs 4GB?
Score
2
June 3, 2012 3:00:03 PM

noob2222take a small enough sample to make sure the results don't prove a thing. why not use some programs that do respond to memory speed?

this IS an article about bulldozer
Score
1
June 3, 2012 4:20:49 PM

SuperVelocelol 100%? Not everyone is only surfing/playing games. And even gamers... not everyone plays badly written games, which can only use one or two cores.P4 was slower because of its extra-long pipelines compared at the same clock, not at half speed. lol


Actually, you're wrong. The pipelines were long, but they were double-pumped, and in any rate, when they went from 20 to 31 stages (Prescott) the performance was pretty much the same (due to other enhancements).

The vast majority of the people who think they know spit out this as the reason, but the main reason was the trace cache was only about 50% effective, and because only one instruction per cycle could be decoded into the trace cache, 50% of the time the processor was running as a scalar processor. When it ran out of the Trace Cache, the performance was very good, but when it didn't, which was quite often, the performance was abysmal.
Score
-1
!