Intel P35 Memory Performance: A Closer Look
by Wesley Fink on May 18, 2007 12:00 PM EST- Posted in
- CPUs
Bandwidth and Memory Scaling
The results chart was greatly expanded for these benchmarks. Instead of grouping together similar timings for easier comparison we have listed every timing used for DDR3 and DDR2 benchmarking. The bold values are the best performing results for that memory speed. We did take a bit of liberty here since the 800 5-6-6 and 6-6-6 timings are considered together, as are the 1066 5-6-6, 6-6-6, and 7-7-7 timings. If that is too broad a consideration for your liking the data is here for you to interpret how you see fit.
The percentages in parentheses after the upper values compare that result to the performance of the equivalent setting on the P965 ASUS P5B Deluxe. This is compared on P965, P35 DDR2, and P35 DDR3 at 1066FSB. As explained on the previous page, 1333 memory speed could not be tested at 1066 CPU FSB which is why that result is missing from the 1066 FSB list.
Results are also present for the P35 boards at 1333 processor bus at the same CPU speed (both at 2.66 GHz as detailed). Again, the limitations of the P965 did not allow us to run comparable 1333 FSB results since the memory ratios are not available for a 1333 FSB CPU.
The 10x266 results are roughly representative of the improvement in memory bandwidth that can be attributed to the memory controller of the P35. Considering the variables in timing, this ranges from around 3% to 6.5%. Since the DDR-800 results are the most comparable across all the memory types, and that ranges from 4.7% to 6.5% improved bandwidth, we would conclude that the contribution of the P35 memory controller to improved memory bandwidth is roughly +5%.
The 2.66GHz 8x333 results show the total impact of the improved P35 memory controller and the increase in processor bus to 1333. As you can see the total improvement in memory bandwidth is 16% to 18%. This means the increase in processor bus speed to 1333 is contributing about 11% to 13% to the bandwidth increase, since we concluded the impact of the memory controller alone to be about 5%.
Those of you who are now saying, "Just as I suspected, it's just the 1333 processor bus that is boosting performance" may want to reserve judgment until the game tests. You will likely be surprised.
We normally also test memory with buffering schemes like MMX, SSE, SSE2, SSE3, etc, turned off. While these features do provide apparent improved bandwidth, we have found the unbuffered bandwidth to correlate better with real-world application performance. Unbuffered performance does not always follow the patterns of buffered memory performance.
The expanded unbuffered results are somewhat different than the unbuffered results in our earlier review. The memory controller contributes from 2% to 8.7% of the memory bandwidth improvement. With our target of the DDR-800 results as most comparable the contribution to improved unbuffered bandwidth is 5%, which is about the same as the buffered contribution. This is pretty unusual, since as you can see below the 1333 line, unbuffered bandwidth improvement is 4% to about 12%, which is lower than the buffered total improvement. For buffered memory bandwidth the memory controller is contributing about 5% of the improvement and the increased processor bus is adding 0% to 6%. Since unbuffered results normally correlate well with gaming performance this hints at the somewhat unexpected gaming results.
The results chart was greatly expanded for these benchmarks. Instead of grouping together similar timings for easier comparison we have listed every timing used for DDR3 and DDR2 benchmarking. The bold values are the best performing results for that memory speed. We did take a bit of liberty here since the 800 5-6-6 and 6-6-6 timings are considered together, as are the 1066 5-6-6, 6-6-6, and 7-7-7 timings. If that is too broad a consideration for your liking the data is here for you to interpret how you see fit.
The percentages in parentheses after the upper values compare that result to the performance of the equivalent setting on the P965 ASUS P5B Deluxe. This is compared on P965, P35 DDR2, and P35 DDR3 at 1066FSB. As explained on the previous page, 1333 memory speed could not be tested at 1066 CPU FSB which is why that result is missing from the 1066 FSB list.
Results are also present for the P35 boards at 1333 processor bus at the same CPU speed (both at 2.66 GHz as detailed). Again, the limitations of the P965 did not allow us to run comparable 1333 FSB results since the memory ratios are not available for a 1333 FSB CPU.
| Standard (Buffered) Sandra XI.SP2 Memory Bandwidth - 2.66GHz | |||
| Memory Speed | P965 ASUS P5B Dlx |
P35 DDR2 ASUS P5K Dlx |
P35 DDR3 ASUS P5K3 Dlx |
| 10 x 266 - 1066FSB - 2.66GHz | |||
| DDR2-800 3-3-3-9 | 5531 | 5754 (+4.0%) |
- |
| DDR2-800 5-6-6-15 | 5207 | - | - |
| DDR2-800 6-6-6-15 | - | 5545 (+6.5%) |
- |
| DDR3-800 6-6-6-15 | - | - | 5451 (+4.7%) |
| DDR2-1067 4-4-3-11 | 5782 | 6037 (+4.4%) |
- |
| DDR2-1067 5-6-6-15 | 5712 | - | - |
| DDR2-1067 6-6-6-15 | - | 5872 (+2.8%) |
- |
| DDR3-1067 7-7-7-20 | - | - | 5843 (+2.3%) |
| 8x333 - 1333FSB - 2.66GHz | |||
| DDR2-800 3-3-3-9 | - | 6456 (+16.7%) |
- |
| DDR2-800 6-6-6-15 | - | 6143 (+18.0%) |
- |
| DDR3-800 6-6-6-15 | - | - | 6156 (+18.2%) |
| DDR2-1067 4-4-3-11 | - | 6811 (+17.8%) |
- |
| DDR2-1067 6-6-6-15 | - | 6621 (+15.9%) |
- |
| DDR3-1067 7-7-7-20 | - | - | 6613 (+15.8%) |
| DDR3-1333 9-9-9-25 | - | - | 6757 |
The 10x266 results are roughly representative of the improvement in memory bandwidth that can be attributed to the memory controller of the P35. Considering the variables in timing, this ranges from around 3% to 6.5%. Since the DDR-800 results are the most comparable across all the memory types, and that ranges from 4.7% to 6.5% improved bandwidth, we would conclude that the contribution of the P35 memory controller to improved memory bandwidth is roughly +5%.
The 2.66GHz 8x333 results show the total impact of the improved P35 memory controller and the increase in processor bus to 1333. As you can see the total improvement in memory bandwidth is 16% to 18%. This means the increase in processor bus speed to 1333 is contributing about 11% to 13% to the bandwidth increase, since we concluded the impact of the memory controller alone to be about 5%.
Those of you who are now saying, "Just as I suspected, it's just the 1333 processor bus that is boosting performance" may want to reserve judgment until the game tests. You will likely be surprised.
We normally also test memory with buffering schemes like MMX, SSE, SSE2, SSE3, etc, turned off. While these features do provide apparent improved bandwidth, we have found the unbuffered bandwidth to correlate better with real-world application performance. Unbuffered performance does not always follow the patterns of buffered memory performance.
| Unbuffered Sandra XI.SP2 Memory Bandwidth - 2.66GHz | |||
| Memory Speed | P965 ASUS P5B Dlx |
P35 DDR2 ASUS P5K Dlx |
P35 DDR3 ASUS P5K3 Dlx |
| 10 x 266 - 1066 FSB - 2.66GHz | |||
| DDR2-800 3-3-3-9 | 4226 | 4370 (+3.4%) |
- |
| DDR2-800 5-6-6-15 | 3668 | - | |
| DDR2-800 6-6-6-15 | - | 3853 (+5.0%) |
- |
| DDR3-800 6-6-6-15 | - | - | 3988 (+8.7%) |
| DDR2-1067 4-4-3-11 | 4608 | 4706 (+2.1%) |
- |
| DDR2-1067 5-6-6-15 | 4389 | - | - |
| DDR2-1067 6-6-6-15 | - | 4484 (+2.2%) |
|
| DDR3-1067 7-7-7-20 | - | - | 4311 -1.8%) |
| 8x333 - 1333 FSB - 2.66GHz | |||
| DDR2-800 3-3-3-9 | - | 4536 (+7.3%) |
- |
| DDR2-800 6-6-6-15 | - | 3975 (+8.4%) |
- |
| DDR3-800 6-6-6-15 | - | - | 4098 (+11.7%) |
| DDR2-1067 4-4-3-11 | - | 4926 (+6.9%) |
- |
| DDR2-1067 6-6-6-15 | - | 4557 (+3.8%) |
- |
| DDR3-1067 7-7-7-20 | - | - | 4547 (+3.6%) |
| DDR3-1333 9-9-9-25 | - | - | 4702 |
The expanded unbuffered results are somewhat different than the unbuffered results in our earlier review. The memory controller contributes from 2% to 8.7% of the memory bandwidth improvement. With our target of the DDR-800 results as most comparable the contribution to improved unbuffered bandwidth is 5%, which is about the same as the buffered contribution. This is pretty unusual, since as you can see below the 1333 line, unbuffered bandwidth improvement is 4% to about 12%, which is lower than the buffered total improvement. For buffered memory bandwidth the memory controller is contributing about 5% of the improvement and the increased processor bus is adding 0% to 6%. Since unbuffered results normally correlate well with gaming performance this hints at the somewhat unexpected gaming results.
Facebook
Twitter
RSS
14 Comments
View All Comments
TA152H - Friday, May 18, 2007 - link
I've looked over these numbers a few times, and tried to make some sense of it, and the remarks about the x6800 being slower on SuperPi, but faster on other things.There were a few clear patterns, which SuperPi managed to break, which leads me to a conclusion that this benchmark needs to be looked at closer.
For example, I noticed that DDR3 shows a greater advantage at the 1333 FSB setting, and starts leaving DDR2 behind. Except on SuperPi.
In every benchmark, 1333 FSB is faster. Except in SuperPi.
The x6800 is faster than other processors on benchmarks at the same settings, except on SuperPi.
Ummmm, anyone else think that there might be something amuck with SuperPi and it probably should be looked at more closely and possibly be removed as a benchmark until it is KNOWN to work correctly. It seems completely wrong. However, if it is reporting correctly, it would be very informative to know why it runs exactly opposite everything else. Is it using the processor in an unusual way? If so, what other applications do, so people know when to pay attention to it, and when not to.
I think the most interesting thing is how the DDR3 likes the 1333 FSB a lot more than DDR2 does. It's clearly the memory of the future.
noobzter - Friday, May 18, 2007 - link
Wes, could you elaborate on that? I've always wondered whether there's any difference between running C2Ds at higher multiplier and higher FSB (e.g. 13x267 vs 10x347)
Wesley Fink - Friday, May 18, 2007 - link
Several editors have noticed that the X6800 at the exact same speed and timing settings is a bit faster in gaming performance and a bit slower in Super Pi performance compared to other C2D processors. This is somewhat contradictory, but .We really don't know why this is the case, but we have seen it over a number of boards and in quite a few test configurations. While we don't know exactly what is different about the X6800 Extreme to make it perform this way, we do know the X6800 Extreme behaves a littel differently in benchmarks. We've asked Intel, but we have never received an answer that explained these minor differences to us.
IntelUser2000 - Friday, May 18, 2007 - link
I can explain that. When Core microarchitecture based CPUs first came out, Intel talked about being able to optimize the prefetchers for the target segment. Mobile chips would have less aggressive and power optimized prefetchers, Desktop chips would be performance optimized(relative to the mobile) and the Xeon variants would be fit for the server/workstation workloads. I would assume the "X" chips could be optimized for gaming, and that may explain the difference.
How much faster is it btw??
Wesley Fink - Saturday, May 19, 2007 - link
If you compare the 1333 gaming numbers from the ddr3 vs. ddr2 review to those in this review you will see the exact difference at the various settings. The ddr3 vs. ddr2 were run with an E6420 at 8x333, while this review used the X6800 at 8x333.Deusfaux - Friday, May 18, 2007 - link
I would guess it would have to do something with cherry picking those particular cores in the fab process and thus getting you the procs from the top of the heap that are just a wee bit better at everything and why they can also justify selling it for such a markup.yacoub - Friday, May 18, 2007 - link
I'd love to know how much performance can be gained on 650i-SLI board going from 10x266 to 8x333. I currently run 10x266 E4400 @ 2.67GHz. I wonder if I'd gain any performance running it at 8x333 and if it'd be worth the added stress to the motherboard to run at 1333MHz fsb. It's an MSi P6N-SLI Platinum.yacoub - Friday, May 18, 2007 - link
Also do I leave my PC6400 DDR2 RAM at 800MHz default or raise it a bit (they don't do 1066 very well), but I could lower them to 667MHz if for some reason it was more ideal to run a 1:2 multiplier from the 333MHz fsb than leaving them up at 800MHz.vailr - Friday, May 18, 2007 - link
Why don't you use the latest beta version Intel chipset drivers?[Instead of: System Platform Drivers Intel - 8.3.0.1013]
Version 8.4.0.1010 Beta:
http://www.station-drivers.com/telechargement/inte...">http://www.station-drivers.com/telechargement/inte...
Wesley Fink - Friday, May 18, 2007 - link
Intel tells us there are no performance imnprovements in the beta drivers. The beta drivers fix a few compatibility issues with Vista. We prefer to use release drivers unless there is a compelling performance or new release reason to use beta drivers. Gary is not testing with the new beta drivers either.