Overclock.net banner

Northbridge speed and low RAM timing significance on AMD platforms

28K views 62 replies 29 participants last post by  vonss 
#1 ·
Moved by luches's recent comparison, I decided to do my own round of comparison on NB speed vs. timings.

The rules/details of this article:
-Timings change between 6-8-6 and 9-9-9
-Northbridge speed changes between 2700Mhz and 3000Mhz
-Memory speed is constant 1600Mhz, the most optimal RAM frequency for AMD. This is not a RAM speed vs. timings comparison. There are no variances in this to keep everything fair.
-Everything else (bus clock, CPU clock, etc.) is also constant.
-The program used to test is "MaxxMem".

So, 20 minutes later:
1600Mhz 6-8-6, 2700Mhz northbridge:
27k6861final.png

1600Mhz 6-8-6, 3000Mhz northbridge:
3k6861final.png

1600Mhz 9-9-9, 2700Mhz northbridge:
27k9991final.png

1600Mhz 9-9-9, 3000Mhz northbridge:
3k9991final.png


1. Does a higher NB provide a benefit?
The answer is yes. You can see in both cases there's about a 500-600MB/s improvement when considering NB speed. Increasing NB speed is one of the ways to provide the biggest performance increases on AMD platforms.
2. Does higher NB speed provide more benefits over tighter timings?
Looks like the answer is also yes. At 3000Mhz 1600 9-9-9 I could whip up better MaxxMem results than at 2700Mhz 1600 6-8-6, so NB speed provides more benefits; of course I couldn't really prove what is equal and what is not (i.e. NB is 300Mhz higher is equal to how many timings lower?) but just so you know it can provide somewhat more benefit. Of course, do consider that you have 5-6ns more latency with CL9 vs CL6-8-6.
3. Do tighter timings still provide benefits?
Of course they will still provide good benefits on AMD. You can see the tighter timings provide a 400-500MB/s benefit in the benchmarks, and the latency is much lower.
4. So, in essence, NB speed and tighter timings scale about the same way?
Yes, so if you are running 1600 CL6 @ 2700-2800NB vs 1600 CL9 @ 3000Mhz NB, you get similar results +/- 100-200MB/s due to similar scaling patterns.
5. Which scales slightly better?
NB speed would scale slightly better over RAM timings/speed. So if you can get a higher NB speed but will have to slightly sacrifice RAM speed/timings, chances are you will get better bandwidth from this, but your latency will suffer.
6. Latency vs. bandwidth? Which is better?
I guess it really depends on what you do. On bandwidth intensive apps such as video editing, certain benchmarks, if you have an SSD, etc. the extra bandwidth provided would probably prove to be a lot better. However when you consider online gaming/ping, boot/app load times, etc. latencies would probably have more effect if the bandwidth is already fairly high. Remember, lowering timings can result in both extra bandwidth and latency. You may also want to look at individual values between different timings/NB speeds, i.e. read bandwidth vs. write bandwidth, depending on what you do.
7. Now what?
In conclusion both are significant and you should go for the highest NB possible and the lowest timings possible, because both give you the greatest results. Of course, lots of memory out there won't reach 1600 @ lower timings than just CL9, and lots of CPU IMCs out there (pretty much everything non-x6) won't reach 3000NB. So if you're looking for absolute best memory performance, these are the things to consider.

Notes:
-CPU speed appears to be around 1200Mhz, this is due to Cool'n'Quiet, something I can successfully apply to my CPU despite overclock.
-Actual CPU speed (during load) is 4.05Ghz in these tests.
-Tests were done on Windows 7 64-bit, on my "Benchmarking" user account which is separate from my main-use user account in that it does not have any startup programs, basic GUI, has more services disabled... basically, more suited for benchmarking.
 
See less See more
4
#3 ·
Thanks, I figured I'd create another thing worth stickying and maybe even prove some things for myself while I'm at it. I'm updating with more answers as I go, please tell me if I've missed anything
wink.gif
 
  • Rep+
Reactions: MightyMission
#4 ·
Excellent write up.i shall re read this again later to ensure it soaks in as i really struggle with ram and timings,i know this could sound silly ,but how do you get to cr1,is it down to performance ram or can you just enter it in the timings in the bios?and is it worth it if you have already caned the nb so much there is little space for any tighter timings?
 
#5 ·
Don't expect 1T command rate with most DDR2, it usually results in a small performance boost. Higher NB won't hold back or affect timings, unless you take the RAM speed up as well.
 
#6 ·
Ahh yes because like i said i dont fully understand ram just yet,
so i let the ram speed rise and the latency fall of course with the increased nb and just up the ram volts,which is very lazy i know...
After re-reading this i will definitely have a tinker tomorrow with the timings and see what it gains.Cheers!
 
#7 ·
Nice work XD, very helpful indeed +1,

Now I gotta work out how to get a higher NB OC on my asrock 890GX extreme 3.

I'm unable to change CPU NB voltage in my bios and can only change NB voltage.

The best I can seem to hit is about 2600Mhz on 1.4 NB Volt
(At about 3.6Ghz CPU clock = fail)

Any advice welcome

(sorry to invade your thread with a question XD, lol)
 
#8 ·
Maybe try a different BIOS? (i.e. newer one). I don't mind questions here either, I answer a lot of them as it is already anyway
tongue.gif
NB voltage is chipset voltage, don't believe it has anything to do with the NB/IMC on the CPU (which is controlled by the CPU-NB voltage).
 
#9 ·
Very nice XD! Always appreciating reading your great RAM info.
 
#12 ·
Got the new bios for my asrock (2.40 is now 2.50)

there was a 2.6 but not on the asrock site and i dont want to mess around with betas.

PC just booted with 2700Mhz NB hope it's stable have to test!

CPU voltages looking better too, might be able to have 4Ghz in summer now we'll see.
 
#14 ·
Must be your RAM binning? What are the rated settings, also what memory voltage are you running at?
 
#16 ·
Well currently I'm running p95 for about an hour on

3300 CPU
2200 NB
2000 HT
1600 RAM @6-8-6-24-33-1T

So far the highest I tried on CL7 was about 2740.

I downloaded superpi and maxxmem2, I will have to compare them with the next thing I will try, 2400 CL6, if this p95 works overnight for 10-12h.
 
#18 ·
So I'm pretty sure some of you have noticed this thread has been stickied now lurks around at the top of the AMD Memory section. I made a few modifications and decided to sticky this up, to perhaps give users some insight on what benefits what, and how significantly it benefits that
smile.gif
 
#20 ·
With my old Asrock 880G extreme3 i was able to set the nb to 2800mhz with 1.4v(but couldnt run the dominators at cl7)-Do not ever buy 880G extreme3
Now i have Asus M4A89GTD-PRO i am able to set the nb to 3000mhz with 1.45v
I Have Corsair Dominator DDR3 1600mhz Cl8 That runs at 1600mhz 7-8-7-20-1T at 1.7v
And there is another thing,there is a latency hit between cpu speed 3200 & 4000,i mean when i Oc My CPu to 4000mhz the latency is lower-so the best perf. opt for me is Cpu at 4000mhz,Nb at 3000mhz,Ram at 1600mhz(7-8-7-20-1)
Cpu-Z Validation :
http://valid.canardpc.com/show_oc.php?id=1825860
An some screen to proof my words:
http://www.overclock.net/picture.php...ictureid=25130
http://www.overclock.net/picture.php...ictureid=25131
I've noticed that when the cpu speed is icreased the mem latency drops
Why Everest Latency and MaxxMem2 Latency scores are diffrent?

Damn my English Suck's

Damn how to upload full res. picture's here?

xd_1771 What happened with your Hexacore?Just Curious.........
 
#21 ·
Still there, just have another PC in sig
running recent benchies for Memory Madness it does seem that CPU speed has some effect on latency, but it is not extremely significant.
 
#26 ·
The CPU sped can open up some more processing of incoming data indeed, but not by any significant amount; while it has been stated that Sandra apparently does not see huge benefit in latency/etc from CPU speed, I have found that the benefit is fairly suitable for benchmarking.

Drop the HT multiplier in the BIOS, if you are able to control it separately from the NB multiplier (you should be).
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top