Memory Latency Impact on Performance
We just looked at the impact of memory bandwidth on performance, but what about latency? Let's first by adjusting the CAS latency from our default of 2 clocks up to 3 clocks. Almost all DDR400 these days is CAS 2 memory, but older memory may have a higher CAS latency or you may have to increase your CAS latency when overclocking to gain more memory bandwidth, so what kind of a performance hit is there when going from CAS 2 to CAS 3?
at_canals_08 |
at_coast_05 |
at_coast_12 |
at_prison_05 |
at_c17_12 |
|
Tcl = 2 |
116.12 |
140.43 |
123.37 |
113.69 |
83.15 |
Tcl = 3 |
115.52 |
137.07 |
121.91 |
113.37 |
79.92 |
At worst, CAS 2 memory seems to be about 5% faster than CAS 3 memory when looking at at_c17_12, our most CPU intensive test. While 5% alone isn't anything major, combine that with a number of other performance tweaks and they can definitely begin to add up.
Now let's look at keeping Tcl (CAS latency) fixed at 2 clocks, but vary Trcd timings from 3 up to 6 clocks:
at_canals_08 |
at_coast_05 |
at_coast_12 |
at_prison_05 |
at_c17_12 |
|
Trcd = 2 |
116.12 |
140.43 |
123.37 |
113.69 |
83.15 |
Trcd = 3 |
115.71 |
136.99 |
122.46 |
113.08 |
79.97 |
Trcd = 4 |
113.92 |
134.42 |
120.87 |
112.38 |
79.83 |
Trcd = 5 |
113.42 |
131.82 |
119.34 |
114.79 |
79.12 |
Trcd = 6 |
113.23 |
128.26 |
117.56 |
111.15 |
77.4 |
For the most part we saw no real changes when adjusting Trcd, the one exception being at_coast_05 which actually showed a pretty big difference between a Trcd value of 2 and higher latency values.
Next we'll look at adjusting Trp:
at_canals_08 |
at_coast_05 |
at_coast_12 |
at_prison_05 |
at_c17_12 |
|
Trp = 2 |
116.12 |
140.43 |
123.37 |
113.69 |
83.15 |
Trp = 3 |
115.6 |
139.24 |
123.13 |
116.35 |
82.09 |
Trp = 4 |
115.85 |
138.88 |
122.98 |
113.16 |
82.05 |
Trp = 5 |
114.84 |
138 |
122.65 |
112 |
80.98 |
Trp = 6 |
114.5 |
136.95 |
121.96 |
115.61 |
80.95 |
Here we see very little impact on performance.
Putting them all together we can see what the overall impact on using fast DDR400, higher latency DDR400 and extremely high latency DDR400 will be:
at_canals_08 |
at_coast_05 |
at_coast_12 |
at_prison_05 |
at_c17_12 |
|
2-2-2-10 |
116.12 |
140.43 |
123.37 |
113.69 |
83.15 |
3-3-3-10 |
114.47 |
134.11 |
120.64 |
112.62 |
80.56 |
3-6-6-10 |
110.74 |
123.76 |
114.75 |
112.17 |
73.8 |
Our standard 2-2-2-10 memory does actually offer reasonable performance benefits in Half Life 2 compared to DDR400 with higher timings such as 3-3-3-10 or the unrealistically high 3-6-6-10.
First and foremost Half Life 2 does appear to be rather dependent on memory bandwidth, but it is also quite appreciative of low latency memory as well. If you're wondering whether being able to run memory at low timings and high clock speeds is important, when it comes to Half Life 2 performance it is.
68 Comments
View All Comments
Phantronius - Wednesday, January 26, 2005 - link
If your gaming on a laptop, you need help.RockHydra11 - Wednesday, January 26, 2005 - link
Was anyone surprised by the results, or didn't know what the answer would be already? I could make a very educated guess before I even clicked on the link.T8000 - Wednesday, January 26, 2005 - link
This kind of CPU reviewing really makes me wonder if anyone plays with a $700 R850 XT-PE without anti-aliasing, anistrophic filtering and does so at 1280x1024.I mean, if that's all you want, why not save a cool $500 and buy a GF6600GT instead.
It would be nice if someone reviewed CPU scaling at real gaming settings, because the 20% differences created here, may translate in only 5% with real settings, making it unnoticable during gameplay.
Whiskyboy - Wednesday, January 26, 2005 - link
I thought the artilce was a nice return to the feature for feature comparison that a shopper like myself really finds useful. I'm slighlty curious about upcoming technologies but I'm really disintereted in seeing how the brand new toy from AMD or Nvidia performs because I'm not going to suggest paying the ridiculous premiums they charge for the new junk. Seeing the effect that things like memory timings, bandwidth, cpu clock have on performance in a consistent platform make it easier for me to make recommendations to my customers for their systems. I like the Buyer's Guide articles, but in all honesty I want the charts that this article has. If you are suggesting that there should be more articles like this, I agree, but I'm not about to complain about the first article in months that actually made a useful shopper's comparison. Thanks AnandMarlowe - Wednesday, January 26, 2005 - link
I would like to see the cpu scaling done with Intel cpu's too! :)Cybercat - Wednesday, January 26, 2005 - link
The X850XT PE being a PCIe part, how did you use it on Socket 754 CPUs? I've not seen any Socket 754 NF4 boards yet.Aquila76 - Wednesday, January 26, 2005 - link
#32 - Gamers buy PC's in orders of magnitude greater numbers than laptops. Maybe you can run Half-life 2 on your Intel Extreme Graphics, but that's nothing compared to gaming on an A64 with a decent video card and sound.jherber - Wednesday, January 26, 2005 - link
where is the pentium m? MOST OF US BUY LAPTOPS THESE DAYS.REMF - Wednesday, January 26, 2005 - link
roflmao:http://images.anandtech.com/graphs/half%20life%202...
an Athlon64 3200+ @ 2.0GHz gets 112fps
an Athlon64 3000+ @ 1.8GHz gets 104fps
.'. an A64 3100+ @ 1.9GHz would get 108fps
...... the same as a P4 570 running at 3.8GHz, twice the speed!
LOL
bupkus - Wednesday, January 26, 2005 - link
I'm glad to finally see this article. I've been waiting for weeks and beginning to think this article was just "vaporware". ;)As to the following quote:
"If you are stuck with one of those older but still well-performing GPUs, don't bother upgrading your CPU unless it's something slower than a 2.4GHz Pentium 4 - you'd be much better served by waiting and upgrading to dual core later on."
As this was just a tantalizing morsel of things to come, I'm looking forward to the coming weeks.