Benchmark tests were run in MacBench 4.0 and the Motorola compiled version of BYTEMARK. Throughout this review I used the standard convention of showing CPU and Cache speeds (in Mhz) in the format of CPU Speed/Cache speed, so 266/177 would mean a CPU speed of 266 Mhz and a backside cache speed of 177 Mhz.

MacBench and BYTEMARK tests were run at several combinations found to be reliable in two days of extensive use. Remember that each card and system has some tolerance variation, so these speeds may not be attainable in every card or system. The specific CPU chip on each card and your motherboard components and RAM can affect maximum reliable speeds. For this reason I've listed the stock card speed in addition to two other settings I was able to attain with this particular card/system combination.

Notes on Speed Settings:

Note: The speeds documented here are for test purposes only and not a recommendation to others to overclock their card. XLR8 is said to honor the warrany regardless of the speeds run but verify that before you attempt settings over the rated CPU speed, Overclocking is risky and can result in loss of data or hardware failures.

The sample card performed reliably far in excess of the rated speed (266/177) in my particular machine. Although I was concerned at first when the card would not boot with the PowerTower Pro motherboard cache installed, removing the cache dimm solved the problem. After removing the cache, no further problems were seen for the duration of the tests.

I was able to obtain reliable operation with the backside cache set to 1:1 at CPU speeds up to 300mhz (300/300) and with the cache ratio set to 1.5:1, the card ran reliably at a CPU speed of 322 Mhz (322/214.6).

With the cover on the case I ran each of these speeds in heavy application use after the machine had been on for 6 hours. Although this is not a guarantee of continued stability, I was confident enough in the card at 300/300 to optimize (defragment) the hard disk at that setting. I actually ran the card for several hours at 332mhz CPU speed, but after completing MacBench tests, a 45 minute Infini-D 4.0 rendering and the ByteMark test (rebooting between each test) the system would not reboot. I backed off to 322/214.6 and retested for 8 hours again with no problems.

As the scores above show - the best performace was obtained at a CPU speed of 322 Mhz with a 1.5:1 cache speed (214.6 Mhz). This resulted in a very slightly higher CPU score than the 300/300 setting, with much better FPU performance (as expected with a faster CPU speed).

I ran the Motorola complied version of BYTEMARK at the stock 266/177, 300/300 and 322/214 speeds with the following results:

The following is a MacBench graph of how this card compared to other G3 CPU cards/systems and the base PowerTower Pro 180 system:

Since the XLR8 card posted the best scores of any card I've tested at the time of this review (March 98), I rated Benchmark performance a 9. (I have to leave some room for the next generation ;-)

The next page has comments and some results from real world applications. Or you may use the links below to jump to a specific page.