3oh6
Well-known member
System Benchmarks
SuperPi Mod v1.5<p style="text-align: justify;"><i>When running the 32M benchmark of SPi, we are calculating Pi to 32 million digits and timing the process. Obviously more CPU power helps in this intense calculation, but the memory sub-system also plays an important role, as does the operating system. SPi 32M has been a favorite amongst benchmarks for these very reasons and is admittedly the favorite benchmark of this reviewer.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-1.png" alt=""></center><p style="text-align: justify;">First out of the gate in the System Benchmarks section is a personal favorite, SuperPi 32M. SuperPi 32M is considered the finesse benchmark amongst enthusiasts because a competitive time takes raw power, but also extensive OS tweaking. Our time is without the OS tweaks so as to concentrate on how the power of the system affects the results. With our almost 22% processor overclock, we are able to shave just over 17% off our 32M time. This means that we aren't quite getting a 1:1 performance gain with 32M. The memory is also overclocked but as we saw in the Everest bandwidth results, that doesn't necessarily mean better as the Copy results were a bit better for the stock setup despite out thoughts that they shouldn't be. Perhaps there is some validity in the Everest Bandwidth results based on our loss of a 1:1 performance gain compared to CPU power increase.</p>
PCMark Vantage<p style="text-align: justify;"><i>The latest iteration of the popular system benchmark is PCMark Vantage from the Futuremark crew. The PCMark series has always been a great way to either test specific areas of a system or to get a general over view of how your system is performing. For our results, we simply run the basic benchmark suite which involves a wide range of tests on all of the sub-systems of the computer.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-2.png" alt=""></center><p style="text-align: justify;">Unlike SuperPi, PCMark Vantage relies on every sub-system of the computer because it tests individual parts specifically then calculates a total score based on the results. Because of this, we don't always see a 1:1 payoff in performance compared to CPU overclock which is clearly the case here. We do, however, get another solid 17% increase in score going from our stock clocks to the overclocked settings. This is almost identical to the gains we saw with the SuperPi 32M calculation. Let's see if this percentage increase continues with a raw CPU benchmark, Cinebench.</p>
Cinebench R10<p style="text-align: justify;"><i>Another benchmarking community favorite, Cinebench renders an intense 2D scene relying on all the processing power it can. Cinebench R10 is another 64-bit capable application and is likely the most efficient program tested today at utilizing all cores of a processor. We will be running both the single threaded and multi-threaded benches here today.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-3.png" alt=""></center><p style="text-align: justify;">And there you have it, the single threaded benchmark of Cinebench saw an over 24% increase in score with the overclocked results while the multi-threaded benchmark saw just under a whopping 28% increase in performance with less than a 22% increase in CPU clocks. We were expecting the numbers to come in around the same percentage but apparently Cinebench doesn't scale perfectly with CPU clocks, it scales better. Synthetic benchmarks are nice, but real life results are better. Let's switch gears and have a look at something useful like encoding time of a DVD to DivX format.</p>
DivX Converter v6.8<p style="text-align: justify;"><i>Next up is a real life benchmark where we simply time a common task done on the computer. Encoding DVDs for viewing on the computer or other devices is an increasingly important task that the personal computer has taken on. We will take a VOB rip of the movie Office Space, and convert it into DivX using the default 720P setting of DivX converter v6.8.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-4.png" alt=""></center><p style="text-align: justify;">In the DivX conversion of the VOB files to a 720p DivX file, we see a drop in time required to encode the video of a hair under 18% or just under seven and a half minutes on a forty one minute conversion. This is very much in line with the previous benchmarks that we just looked at like SuperPi and PCMark Vantage and goes to show that overclocking just isn't for benchmarking.</p>
Lame Front End<p style="text-align: justify;"><i>Un-like the DivX conversion we just looked at, Lame Front End is not multi-threaded and only utilizes a single core of a processor. This will obviously limit performance but we should still recognize significant time savings going from the stock settings to the overclocked results. We will be encoding a WAV rip of the Blackalicious album, Blazing Arrow and converting it to MP3 using the VBR 0 quality preset.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-5.png" alt=""></center><p style="text-align: justify;">The single threaded sister to our real life DivX encoding test, Lame Front End is simply a GUI for encoding WAV files to MP3. Audiophiles still consider Lame - single threaded - to be the best way to encode audio and that likely won't change. Needless to say, the results are exactly what we have come to expect from this round of benchmarks if not slightly lower with the overclocked settings coming in 17% quicker than our stock settings. This equates to about twenty six seconds on a two and a half minute task or a full album.</p>
Photoshop CS4<p style="text-align: justify;"><i>Adobe Photoshop CS4 is fully x64 compliant and ready and able to use every single CPU cycle our processor has available including the implementation of GPU support utilizing the GTX 280 in our test system. We have changed our Photoshop benchmark to more of a standardized test configured by DriverHeaven.net. Their Photoshop benchmark utilizes 12 filters and effects on an uncompressed 60MB .JPG image that will test not only the CPU but also the memory subsystem of our test bench. Each portion of the benchmark is timed and added together for a final time that is compared below.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-6.png" alt=""></center><p style="text-align: justify;">The last of the benchmarks we are going to break down is a new one for us here at HWC. We have decided to go with a bit of an industry standard in Photoshop benchmarks for this review and used the DriverHeaven.net Photoshop benchmark script. This benchmark seems to be a rather accurate depiction of commonly used Photoshop filters and seems to be accepted amongst the review and consumer public as a good Photoshop test. The results appear to agree with these views as the time to complete the script of filters and effects comes in at 18.8% less for the overclocked system versus our stock setup. This, again, is perfectly in-line with what we have seen from almost every benchmark, synthetic or real life today.
Overall, we are quite impressed with what this setup offers in terms of performance in our favorite synthetic benchmarks as well as real life performance. We know that we haven't compared this setup to that of generations past and that is because we feel like we would be simply repeating history in doing so. Our resident comparison guru, Mac, has already put together an extensive collaboration of results comparing the new i7 platform to that of Intel and AMD past and if you were interested in seeing how the i7 platform stacks up, we again encourage you to view our Intel Core i7 "Nehalem" 920, 940 & 965 XE Processor Review. We still have one order of business to cover here though, and that is the 3D and gaming benchmarks including four of the most popular gaming titles out right now.</p>
SuperPi Mod v1.5<p style="text-align: justify;"><i>When running the 32M benchmark of SPi, we are calculating Pi to 32 million digits and timing the process. Obviously more CPU power helps in this intense calculation, but the memory sub-system also plays an important role, as does the operating system. SPi 32M has been a favorite amongst benchmarks for these very reasons and is admittedly the favorite benchmark of this reviewer.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-1.png" alt=""></center><p style="text-align: justify;">First out of the gate in the System Benchmarks section is a personal favorite, SuperPi 32M. SuperPi 32M is considered the finesse benchmark amongst enthusiasts because a competitive time takes raw power, but also extensive OS tweaking. Our time is without the OS tweaks so as to concentrate on how the power of the system affects the results. With our almost 22% processor overclock, we are able to shave just over 17% off our 32M time. This means that we aren't quite getting a 1:1 performance gain with 32M. The memory is also overclocked but as we saw in the Everest bandwidth results, that doesn't necessarily mean better as the Copy results were a bit better for the stock setup despite out thoughts that they shouldn't be. Perhaps there is some validity in the Everest Bandwidth results based on our loss of a 1:1 performance gain compared to CPU power increase.</p>
PCMark Vantage<p style="text-align: justify;"><i>The latest iteration of the popular system benchmark is PCMark Vantage from the Futuremark crew. The PCMark series has always been a great way to either test specific areas of a system or to get a general over view of how your system is performing. For our results, we simply run the basic benchmark suite which involves a wide range of tests on all of the sub-systems of the computer.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-2.png" alt=""></center><p style="text-align: justify;">Unlike SuperPi, PCMark Vantage relies on every sub-system of the computer because it tests individual parts specifically then calculates a total score based on the results. Because of this, we don't always see a 1:1 payoff in performance compared to CPU overclock which is clearly the case here. We do, however, get another solid 17% increase in score going from our stock clocks to the overclocked settings. This is almost identical to the gains we saw with the SuperPi 32M calculation. Let's see if this percentage increase continues with a raw CPU benchmark, Cinebench.</p>
Cinebench R10<p style="text-align: justify;"><i>Another benchmarking community favorite, Cinebench renders an intense 2D scene relying on all the processing power it can. Cinebench R10 is another 64-bit capable application and is likely the most efficient program tested today at utilizing all cores of a processor. We will be running both the single threaded and multi-threaded benches here today.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-3.png" alt=""></center><p style="text-align: justify;">And there you have it, the single threaded benchmark of Cinebench saw an over 24% increase in score with the overclocked results while the multi-threaded benchmark saw just under a whopping 28% increase in performance with less than a 22% increase in CPU clocks. We were expecting the numbers to come in around the same percentage but apparently Cinebench doesn't scale perfectly with CPU clocks, it scales better. Synthetic benchmarks are nice, but real life results are better. Let's switch gears and have a look at something useful like encoding time of a DVD to DivX format.</p>
DivX Converter v6.8<p style="text-align: justify;"><i>Next up is a real life benchmark where we simply time a common task done on the computer. Encoding DVDs for viewing on the computer or other devices is an increasingly important task that the personal computer has taken on. We will take a VOB rip of the movie Office Space, and convert it into DivX using the default 720P setting of DivX converter v6.8.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-4.png" alt=""></center><p style="text-align: justify;">In the DivX conversion of the VOB files to a 720p DivX file, we see a drop in time required to encode the video of a hair under 18% or just under seven and a half minutes on a forty one minute conversion. This is very much in line with the previous benchmarks that we just looked at like SuperPi and PCMark Vantage and goes to show that overclocking just isn't for benchmarking.</p>
Lame Front End<p style="text-align: justify;"><i>Un-like the DivX conversion we just looked at, Lame Front End is not multi-threaded and only utilizes a single core of a processor. This will obviously limit performance but we should still recognize significant time savings going from the stock settings to the overclocked results. We will be encoding a WAV rip of the Blackalicious album, Blazing Arrow and converting it to MP3 using the VBR 0 quality preset.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-5.png" alt=""></center><p style="text-align: justify;">The single threaded sister to our real life DivX encoding test, Lame Front End is simply a GUI for encoding WAV files to MP3. Audiophiles still consider Lame - single threaded - to be the best way to encode audio and that likely won't change. Needless to say, the results are exactly what we have come to expect from this round of benchmarks if not slightly lower with the overclocked settings coming in 17% quicker than our stock settings. This equates to about twenty six seconds on a two and a half minute task or a full album.</p>
Photoshop CS4<p style="text-align: justify;"><i>Adobe Photoshop CS4 is fully x64 compliant and ready and able to use every single CPU cycle our processor has available including the implementation of GPU support utilizing the GTX 280 in our test system. We have changed our Photoshop benchmark to more of a standardized test configured by DriverHeaven.net. Their Photoshop benchmark utilizes 12 filters and effects on an uncompressed 60MB .JPG image that will test not only the CPU but also the memory subsystem of our test bench. Each portion of the benchmark is timed and added together for a final time that is compared below.</i></p><center><img src="http://images.hardwarecanucks.com/image/3oh6/evga/x58sli/sys_bench-6.png" alt=""></center><p style="text-align: justify;">The last of the benchmarks we are going to break down is a new one for us here at HWC. We have decided to go with a bit of an industry standard in Photoshop benchmarks for this review and used the DriverHeaven.net Photoshop benchmark script. This benchmark seems to be a rather accurate depiction of commonly used Photoshop filters and seems to be accepted amongst the review and consumer public as a good Photoshop test. The results appear to agree with these views as the time to complete the script of filters and effects comes in at 18.8% less for the overclocked system versus our stock setup. This, again, is perfectly in-line with what we have seen from almost every benchmark, synthetic or real life today.
Overall, we are quite impressed with what this setup offers in terms of performance in our favorite synthetic benchmarks as well as real life performance. We know that we haven't compared this setup to that of generations past and that is because we feel like we would be simply repeating history in doing so. Our resident comparison guru, Mac, has already put together an extensive collaboration of results comparing the new i7 platform to that of Intel and AMD past and if you were interested in seeing how the i7 platform stacks up, we again encourage you to view our Intel Core i7 "Nehalem" 920, 940 & 965 XE Processor Review. We still have one order of business to cover here though, and that is the 3D and gaming benchmarks including four of the most popular gaming titles out right now.</p>
Last edited by a moderator: