Review: BIOS 1.1 saves the day
Six weeks ago we presented you the new MSI P45 Platinum here, and we thought we can do the benches rather quick. While testing the BIOS we found several strange defaults and some bugs, but we are always in intensive contact with MSI to fix all the issues we find. The folks from MSI in Holland are very very supportive.
Some of the bugs are still present, MSI did not manage to fix some settings and what's worse it was not able to improve oc-capabilities since the release. Last Monday we saw the BIOS 1.1 on their site, the date of June 16th is of course misleading, because last week it was not available. The sad thing is, we were not informed about it. Re-running oc-test is a bit time-consuming and of course delayed this review again.
As usual MSI uses an AMI BIOS. The new thing is, it's 4MB big. This is of course quite disappointing, because MSI still did not manage to build in a flash-tool. Also the Live-Update Tool MSI is providing leaves much to be desired. As long as you use any beta-BIOS it does not detect it and will not give you any opportunity to update.
On the AMI website you will find the AMI Windows Flash-Tool, which showed us something amazing. The BIOS is actually only 1MB, but it seems to be mirrored 4 times, because the flash-tool was writing the same areas four times. The 4MB is needed for future BIOS upgrades to EFI (Extensible Firmware Interface) and MSI is already providing EFI for some of its motherboards.
Besides some awkward settings, for example: COM1 2F8, which is still present in the V1.1 BIOS the most annoying thing is that the "Greenpower" feature is disabled. We got in quite a fight with MSI because we think, if you do advertisments all over computer magazines and websites with their "DrMOS" marketing then you should also enable that option. The guys in Holland understood our point of view, but it seems the Chinese don't know what their marketing department is doing, so we got an unusual reply from them, saying they are afraid that users may use low quality memory and this may result in instability.
Our argument is, if you go for a board which will set you back €140/$200 you will not put in crappy memory. Also Greenpower consists of three settings for CPU, MCH and memory. Most power-saving is done by disabling some phases of the VRM for the CPU, so disabling only the memory portion wouldn't have hurt power-consumption considerably.
We have decided to enable that feature during testing and it worked fine. Not enabling it costs lots of energy, so we urge you to enable it and ignore any statement MSI put out, as long as you don't use noname memory. MSI put a new feature in called "Memory-Z" which gives you all the SPD readings. Many vendors do that already but they put the results in the settings screen. MSI decided to give them an extra page, so you have to jump forward and backwards to apply the readings manually which is also not an elegant way to do things.
We found some rather insane voltage values inside the BIOS, which you should never ever apply:
MCH voltage settings up to 2.6V, do not set it higher than 1.60V, click for bigger screenshot.
Memory voltages up to 3.3V, the highest rated memories do only take 2.3V, with extra cooling you can get up to 2.5V on certain modules. If you don't use nitrogen for memory cooling, don't even think of applying 3V. Click for bigger screenshot.
Frontsidebus Termination Voltage VTT, never ever apply more than 1.40V, apply 1.40V only when using water-cooling. It's recommended to stay at 1.20V because higher voltages increase interference and thus decrease stability. Click for bigger screenshot.
With the V1.0 BIOS overclocking was nearly a complete failure. We could only manage 485MHz FSB, because our CPU is one of the first off the shelves and behaves quite stubbornly when OC is applied.
The V1.1 BIOS lets gets us to 530MHz stable. Some months ago that would have been an impressive result, but meanwhile some boards go up to 600MHz. We are not sure if this a the limit of our CPU, but testing with an new Intel E8500 sample did not improve the score.
The board is not an overclocker's dream, we were stuck at about 4.3GHz, 4.4GHz was never stable.
The board seems to have a limit around 1100MHz. At least with our DDR2 modules even with 2.30V applied we could not get to 1200MHz for stable operation.
We have to state that we always recalculate the bench results to nominal frequencies. Most vendors do an overclock to their products, maybe to get more bench points, but we nullify such attempts. MSI most of the time overclocks and we don't like that. Some settings we reduced intentionally inside the BIOS, for example we used FSB 474 to get 475.
Please note that different bios revisions may give different results. All benches are done with AUTO settings without altering any BIOS option besides CPU VCore, NB VCore and FSB Termination Voltage.
x264 is a h.264/AVC codec which supports four threads, and it's available for free. We took a PAL episode of "Babylon 5" with a length of 41 minutes, 57 seconds and 8 frames. We tried to "emulate" the most common usage when you encode your movies:
1st: We have a perfect master, so we only de-interlace the content and resize it without any other manipulations; we marked this as "fast."
2nd: You get bad mastering on many DVDs, especially "old" stuff or when the studios are in a hurry for the release. In this case you want to improve the picture quality, which is done by filtering the content. You can choose from lots of filters for any purposes you can think of, but we only used the most common "undot," "FluxSmooth" and "MSharpen." Of course, we also de-interlaced, filters were done before any resizing took place (which is slower). We marked this as "slow."
*Note: The DFI P35 mainboard doesn't support half multipliers, so it was using 417x8.
The same episode we encoded, we used for our MP3-testing. We don't recommend using MP3 for encoding, because AC3 can do the job better, but nearly 42 minutes gives us approximately the length of any given album.
A measurement in seconds, as many sites do, is useless, because the differences are too small. So we used the built-in play/CPU ratio, this means the CPU is encoding x-times faster then the track-length. Fast memory does not play an important role here. For your convenience we also show the single-threaded benchmarks figures, they can be re-produced with any version of L.A.M.E. Only LameMT can do multi-thread and take advantage of multi-core processors.
We used this setting: lamemt --vbr-new -q 2 -V 2 -m j --strictly-enforce-ISO --resample 48
*Note: The DFI P35 mainboard doesn't support half multipliers, so it was using 417x8.
With Greenpower enabled the board does shine.
Please note that other configurations may yield other results.
Pushing an unfinished product on the market is never a good move. The board was on sale in early June and it took MSI one month to release a BIOS that would enable a decent overclock.
Most annoyingly MSI decided to disable the energy-savings per default, which is one of the more odd things to do, especially considering MSI's marketing department advertises this feature around the world in important magazines and on the biggest websites. On the other hand, they incorporated some insane voltage settings which for sure will ruin your board and your memory. Besides some BIOS engineers are still in the 90ties where S1 was an accomplishment, but we expect any board to run with S3 settings without trouble. Enabling features such as COM1 which is not on the back-panel or slot-bracket is a waste of resources. We all know, the less resources Windows has to handle the more stable it gets.
The board is a decent product, but it's no overclocker's dream. It seems HQ is putting all effort in their DDR3 line which we will come out soon. DDR3 has become cheaper in the last months but still is twice as expensive compared to DDR2 and does not offer any significant speed improvements. High overclocking 2GB DDR3 memory kit still cost about €180,- while you can get 4GB DDR2 memory kit for about €60,-.
While our discussion with MSI improved their layout which is nearly perfect now, the BIOS is still in need of much more improvement. We would like to see energy savings enabled by default, voltages limited so it will not damage your board and settings enabled/disabled which do make sense. We also think the board was not designed to be an overclockers dream, MSI putting all effort to their DDR3-line, but giving you enough headroom to get decent overclocking and performance.
This board is now available for about €137,- and if you have some knowledge of the BIOS and settings you can buy it without thinking, otherwise you should consider alternatives.