TF536 + Spectre GCR is slower than stock.
Posted: 27 Feb 2021 16:15
Here's a strange thing I decided to experiment with the TF536 and see how much faster MacOS under Spectre GCR was when running on a 68030. Both Spectre GCR and MacOS know about the 68030. Indeed, there are options in Spectre GCR 3.0 to enable the caches and burst mode.
So I first ran Speedometer 3.06 under MacOS System 6.0.7 using the 68000 to get a baseline benchmark. As expected it was about 20% faster than an original Mac:
I then shut down, switch off and swapped over to the TF536 and did the same thing. Speedometer recognised the 68030 in the system info. To my surprise it was actually *SLOWER*.
Here's the comparison table (poor photo):
My only guess is that Spectre GCR is somehow compensating for the CPU speed so as to make sure the system runs at the same speed as the original Mac Plus/512Ke. Very strange.
For reference, which is what I get with my SE/30, which has a lower clocked 68030 (but faster memory).
So I first ran Speedometer 3.06 under MacOS System 6.0.7 using the 68000 to get a baseline benchmark. As expected it was about 20% faster than an original Mac:
I then shut down, switch off and swapped over to the TF536 and did the same thing. Speedometer recognised the 68030 in the system info. To my surprise it was actually *SLOWER*.
Here's the comparison table (poor photo):
My only guess is that Spectre GCR is somehow compensating for the CPU speed so as to make sure the system runs at the same speed as the original Mac Plus/512Ke. Very strange.
For reference, which is what I get with my SE/30, which has a lower clocked 68030 (but faster memory).