deeppow
08-03-07, 09:24 PM
This is a contraction of various parts of a thread posted here (http://www.ocforums.com/showthread.php?t=521047). Thought I would condense and summarize in hopes of helping a few overclockers. Should the admins not find this acceptable, please delete.
Have run some tests to see what are the dominate overclocking parameters for the Core 2 Duo, E6600 in my case. Results were obtained by using rig noted in my sig. Performance was benchmarked using the Pi 1M digits test unless otherwise stated.
You do get better performance with higher CPU speed, see Figure 1, no surprise here of course. Memory speed and timings are held constant in this study. You can also observe that there is minimal effect of the FSB speed when the CPU and memory speeds are fixed, i.e. running the northbridge faster alone doesn't provide any advantage.
Figure 2 shows the gains of higher memory speed at a three different CPU speeds that were 3.5Ghz, 3.0Ghz and 2.4Ghz. Memory timings are held constant in these studies. As you can see, higher memory speed does improve performance. But the difference between memory at 800 and 1000 is minimal, on the order of 1.5% or less (see details below).
Next, here (http://www.deep-powder.net/OC_Guide/Data/Pi_MEMtimings.htm) is info on the effects of memory timings. Interesting detective story associated with the initial part of investigation. The table (http://www.deep-powder.net/OC_Guide/Data/Pi_MEMtimings.htm) shows results using my Mushkin that is rated at 4-5-4-11. Two sets of tests associated with memory timings are shown. One test ran Pi using 1M digits with 5 samples and the other 2M digits with 10 samples. The 1M case is often used as a reference test by a number of folks including myself.
To make a long story short, the results using 1M digits and 5 samples do not produce results that we would expect. For example 5-5-5-12 is faster than 4-4-4-11, very questionable result. However if you look at the standard deviations of the tests (called sigma in table), you see that the variability (as represented by sigma) indicates that while the results aren't what we expect it is a reasonable result. (If you have a collection of data from a Normal Distribution then approximately 66% of the data should fall within one standard deviation of the mean.) Thus 66% of the average values would lie between 14.7689 (average - sigma) and 14.8835 (average + sigma) if I had an infinite number of tests. My result for 5-5-5-12 lies within that spread so I can't say it is wrong. Thus my metric as applied to this question doesn't appear to be good enough! You must always make sure you're measurement reflects what you think it does.
From this point, I conclude I required a better metric to measure the effect of memory timings. For this reason, I conducted the Pi test set with 2M digits and 10 samples. Took half a day testing to run those cases in combo with a few "honey-do"s too. Anyway now looking at those results in the table, they are more typical of what we expect, i.e. the ordering of what is important is typical for the various settings.
So what is the maximum improvement that I might reasonably expect with better memory timings? 66% of my maximum performance improvement will be less than 1% for a single setting change to all timings, i.e. going from 5-5-5-12 to 4-4-4-11. I get this number from this (37.4248+.0.1108)/(37.2565-.08831)=1.0099 or 1%.
How does this relate to better memory? If I had DDR2 rated at 800 and went to memory rated at 1000 using data I've not given you but did show in the figure, (15.019+0.017)/(14.854-0.034)=1.0146 or 1.5%. Thus better (faster) memory is better, in this case 800 at 4-4-4-11 will be better than 1000 at 5-5-5-12. Of course taking bigger steps in memory speed or wider spreads in timings would have to be tested so be careful with extrapolation.
From these 3 sets of studies, I conclude
1) CPU speed is most important and greatest payback
2) higher memory speed is next
3) tighter memory timings (speed and timings usually compete with one another, as speed gets faster the timings get slower)
4) FSB alone has minimal value.
Anyway, those are the basics with the regions I've tested.
EXTRA:
A little more comparative info is shown in Figures 3 and 4. Figure 3 shows performance as a function of CPU speed for two cases, memory run at ~1066 and ~600 which are the extreme memory speeds I’ve tested. As you can see observe at 3Ghz, the performance difference due memory speed is ~3.7% which would require and additional overclock of 100Mhz for the 600 memory to be equivalent. Depending of the cost difference between 600 and 1066, you would have to decide if the performance payback value is worth the extra cost for you.
Figure 4 refines (shows in greater detail) part of a previous plot. From this plot you can observe that if you bought 800 memory, you would get 2/3 of the potential gain between 600 and 1066. Thus 800 might be a good option for you if the price difference for the 1000 speed is too high for you.
Looking at the table noted above, you can see that buying memory with 4-4-4-N timings would give you a gain of ~0.5% in performance over memory with 5-5-5-N timings. Again, you need to decide value.
Have run some tests to see what are the dominate overclocking parameters for the Core 2 Duo, E6600 in my case. Results were obtained by using rig noted in my sig. Performance was benchmarked using the Pi 1M digits test unless otherwise stated.
You do get better performance with higher CPU speed, see Figure 1, no surprise here of course. Memory speed and timings are held constant in this study. You can also observe that there is minimal effect of the FSB speed when the CPU and memory speeds are fixed, i.e. running the northbridge faster alone doesn't provide any advantage.
Figure 2 shows the gains of higher memory speed at a three different CPU speeds that were 3.5Ghz, 3.0Ghz and 2.4Ghz. Memory timings are held constant in these studies. As you can see, higher memory speed does improve performance. But the difference between memory at 800 and 1000 is minimal, on the order of 1.5% or less (see details below).
Next, here (http://www.deep-powder.net/OC_Guide/Data/Pi_MEMtimings.htm) is info on the effects of memory timings. Interesting detective story associated with the initial part of investigation. The table (http://www.deep-powder.net/OC_Guide/Data/Pi_MEMtimings.htm) shows results using my Mushkin that is rated at 4-5-4-11. Two sets of tests associated with memory timings are shown. One test ran Pi using 1M digits with 5 samples and the other 2M digits with 10 samples. The 1M case is often used as a reference test by a number of folks including myself.
To make a long story short, the results using 1M digits and 5 samples do not produce results that we would expect. For example 5-5-5-12 is faster than 4-4-4-11, very questionable result. However if you look at the standard deviations of the tests (called sigma in table), you see that the variability (as represented by sigma) indicates that while the results aren't what we expect it is a reasonable result. (If you have a collection of data from a Normal Distribution then approximately 66% of the data should fall within one standard deviation of the mean.) Thus 66% of the average values would lie between 14.7689 (average - sigma) and 14.8835 (average + sigma) if I had an infinite number of tests. My result for 5-5-5-12 lies within that spread so I can't say it is wrong. Thus my metric as applied to this question doesn't appear to be good enough! You must always make sure you're measurement reflects what you think it does.
From this point, I conclude I required a better metric to measure the effect of memory timings. For this reason, I conducted the Pi test set with 2M digits and 10 samples. Took half a day testing to run those cases in combo with a few "honey-do"s too. Anyway now looking at those results in the table, they are more typical of what we expect, i.e. the ordering of what is important is typical for the various settings.
So what is the maximum improvement that I might reasonably expect with better memory timings? 66% of my maximum performance improvement will be less than 1% for a single setting change to all timings, i.e. going from 5-5-5-12 to 4-4-4-11. I get this number from this (37.4248+.0.1108)/(37.2565-.08831)=1.0099 or 1%.
How does this relate to better memory? If I had DDR2 rated at 800 and went to memory rated at 1000 using data I've not given you but did show in the figure, (15.019+0.017)/(14.854-0.034)=1.0146 or 1.5%. Thus better (faster) memory is better, in this case 800 at 4-4-4-11 will be better than 1000 at 5-5-5-12. Of course taking bigger steps in memory speed or wider spreads in timings would have to be tested so be careful with extrapolation.
From these 3 sets of studies, I conclude
1) CPU speed is most important and greatest payback
2) higher memory speed is next
3) tighter memory timings (speed and timings usually compete with one another, as speed gets faster the timings get slower)
4) FSB alone has minimal value.
Anyway, those are the basics with the regions I've tested.
EXTRA:
A little more comparative info is shown in Figures 3 and 4. Figure 3 shows performance as a function of CPU speed for two cases, memory run at ~1066 and ~600 which are the extreme memory speeds I’ve tested. As you can see observe at 3Ghz, the performance difference due memory speed is ~3.7% which would require and additional overclock of 100Mhz for the 600 memory to be equivalent. Depending of the cost difference between 600 and 1066, you would have to decide if the performance payback value is worth the extra cost for you.
Figure 4 refines (shows in greater detail) part of a previous plot. From this plot you can observe that if you bought 800 memory, you would get 2/3 of the potential gain between 600 and 1066. Thus 800 might be a good option for you if the price difference for the 1000 speed is too high for you.
Looking at the table noted above, you can see that buying memory with 4-4-4-N timings would give you a gain of ~0.5% in performance over memory with 5-5-5-N timings. Again, you need to decide value.