Biostar TForce 6100 Motherboard
Author: Zahn Funk
Editor: Shawn Knight
Date: 06-13-2006
Pages:
Final Setup

The last setting available in the ONE menu is for the MIT feature, or Memory Integration Test.



Enabling this will run a version of the Memtest86 utility on next boot. Overclockers familiar with this utility know how invaluable it is when testing out new hardware. I found it very useful to have this utility built-in as I don't have a floppy drive attached to this system.



Other features available from the main BIOS screen are the CRP and IFP. The CMOS Reload Program allows you to store up to 50 instances of saved BIOS settings and reload them with ease. If the BIOS settings are ever lost or reset, rather than loading optimized settings and then have to go through each menu and make changes, you can simply reload from a previously saved session. The Integrated Flash Program can be used to upgrade the BIOS version, although it still requires you to have the actual bin file to be flashed available on floppy.



The last BIOS feature, SRS or Self Recovery System, is not a selectable menu or option. What it refers to is the board's ability to recover from a failed POST in the event you've made some change to BIOS that the system did not like. After a few seconds of black screen the TForce will reset and return to the main BIOS screen. From that point you can use the CRP to reload your last working settings (you did save them didn't you?) and very quickly be back in operation.

The Driver/Software CD contains a few utilities that can be used from within Windows. One of these is the TForce Hardware Monitor. This program can be set to launch on boot and minimize to the task bar. It monitors CPU temperature, fan speeds and voltages, and will alert the user if any of these variables fall out of spec. The allowable range can be manually adjusted from within the utility.



The TForce Overclocking utility is supposed to allow the user to change CPU and memory frequencies as well as voltages from within Windows. Unfortunately every time I launched the program I was greeted with a corrupt display and a hung system. I've had similar things happen with other in-Windows overclocking utilities before so it didn't really surprise me. I'd rather trust my overclocking to the BIOS anyway.

Speaking of overclocking, after loading the BIOS defaults and tweaking the settings I started out by jumping the CPU frequency to 250, then began increasing that in 10Mhz increments, using the built-in Memtest86 utility to verify stability at each step. I soon reached the maximum speed of the memory (with 2.9v) at 290, and had to introduce the 166 divider. After just one more increase, at 300 exactly, the system refused to go any higher.



The 2500+ Sempron 64 is clocked by default at 200 x 7 for a resulting 1400Mhz. With the 300 frequency setting the CPU was posting at a respectable 2100Mhz, and I hadn't had to increase the CPU voltage at all. Strangely, any further would result in failed POSTs or Memtest errors, no matter if I increased the CPU or DDR voltages or dropped the HT Multi or Mem divider. Although a 700Mhz increase is nothing to sneeze at, I found it odd to hit a brick wall that apparently could not be overcome. Since the default (and thus highest) multiplier available on the 2500+ is 7x, it is not possible to go any higher on the CPU frequency.

With the 166 memory divider, the DDR frequency is reduced to 233, and even after lowering the timings to 2.5-3-3-7 system performance was still worse than at the highest 1:1 setting. Therefore I lowered the CPU frequency to 290 and put the memory back on 200 with 2.9v and 3-4-4-8 timings. This results in the CPU running at 2030Mhz.



On to testing and conclusion.


  Sponsors Links

  Sponsors Links

  Newsletter

Submit your email address to subscribe to the OCIA.net Newsletter for the latest site updates!

Email Address: