With the receipt of our first true Super 7 motherboard, the Soyo SY-5EH, we set about the task of retesting our batch of test CPUs to find out whether a Super 7 board could indeed run any of the current CPUs at 100MHz. What we found was that there are more current processors able to handle the 100MHz bus speed than we thought, however there are still a few that cannot.
The processors we tested are a fairly representative sample of what has been available in the past few years, and should be indicative of the types of processors currently in use. That should make this article of keen interest to a very wide audience. We set the jumpers and switches on the SY-5EH to run the system bus, the Host Bus and SDRAM all at 100MHz. The board allows each of these to be set independently, so it is possible that some tweaking of these will allow the CPU to run at 100MHz, while keeping, for example, the SDRAM at 66MHz for compatibility reasons.
Here is the list of processors we tested, and the results of whether they would run at a 100MHz bus speed…
IDT C6 225MHz | 2x100MHz | No |
Cyrix 6×86 PR200 | 2x100MHz | No |
Cyrix 6x86L PR200 | 2x100MHz | No |
Cyrix 6x86MX PR200 | 2x100MHz | No |
Cyrix 6x86MX PR233 | 2x100MHz | No |
AMD K6-200 | 2x100MHz | Yes |
AMD K6-233 | 2x100MHz | Yes |
AMD K6-300 | 3x100MHz | Yes |
Intel P54C 150 | 1.5x100MHz | Yes |
Intel P55C 166 | 2x100MHz | Yes |
Intel P55C 233 | 2x100MHz | Yes |
The most interesting results are that none of the Cyrix processors were able to run at 100MHz. The 6x86MX PR233 came the closest, and with some tweaking and fooling with the jumpers on the SY-5EH, we might have been able to get it to run. The other result that surprised us was that the Classic Pentium P150 ran without a problem at 100MHz.
The tests we ran with the R581A seemed to indicate that only the AMD and Cyrix PR233 processors would be able to run at 100MHz, but this obviously was not correct. It still looks like the CPU does have some determination as to whether the processor will be able to work at 100MHz, but the chipset and/or cache speed has a bigger role than we expected to see. Obviously, the L1 cache wasn’t a factor at all here.
Beyond 100MHz
After testing these processors, we noticed that the SY-5EH has an even faster setting for the bus speed – 112MHz! This setting will run the PCI bus at about 37.3MHz and the AGP bus at about 75MHz, which is almost the same as running at the synchronous 75MHz setting that many people have tried.
Using Micron PC100 SDRAM and an Intel Pentium 166MMX processor, we set the board to run at 112MHz x 2.0, and to our surprise, the board reported a Pentium MMX @ 225MHz, then proceeded to boot into Windows and run applications for about 2 hours before we decided to shut it down. No crashes, hangs, etc.
This now begs the question about what the upper limit of the Socket 7 processors might be when used with the proper chipset. 125MHz? 133MHz? This should provide some interesting fodder for thought over the next several months as motherboard manufacturers gain a better understanding of the capabilities of various chipsets and how to best implement them.
Without having any other Super 7 boards to ‘play’ with at this time, we don’t know if other motherboards will have the 112MHz or faster bus speeds, but Soyo appears to have done a spectacular job in implementing this speed on their board. One item to note is that while the chipset says ETEQ, it is really a VIA MVP3 chipset under the covers. It seems that VIA has come through with flying colors on this chipset, considering the other two major manufacturers (ALi and SiS) have not yet successfully produced a chipset that will run faster than 90MHz in a reliable fashion.
Be the first to discuss this article!