As I understand the way the Super Game Boy worked, it relied on a derivative of the SNES's main clock, so it runs at 4.295454 MHz instead of 4.194304 MHz. In other words, compared to a handheld Game Boy, the SGB is overclocked. The number of clock cycles to display the screen does not change, it remains at 70,224. Thus the SGB's refresh rate is 61.167 Hz. Despite the 2.4% clock difference, I have never seen any visual glitching on an SGB that I did not see on a DMG or GBP.
The Super Game Boy 2 has its own clock, a proper 4.194304 MHz. The SGB2's refresh rate should be 59.7275 Hz, but the sketchy sources I have seen suggest it is the SNES's 60.0988 MHz. I guess the theory is that a cartridge can drive the SNES faster than 60.0988 Hz but not slower. You can also mod an SGB with the proper clock crystal to run at the canonical 4.194304 MHz. I would hope it would behave identically in terms of framerate, to the SGB2.
I recall that someone on this board indicated that there may have been some benefit to the SGB's method beside being cheaper to make. If there is a discrepancy between the SGB2's framerate output and the SNES's framerate output, then that would suggest that there must be some frame-skipping going on that would not be present with an unmodded SGB.
Since I now own an SGB and an SGB2, I took video captures of a game scrolling left and right. The SGB's capture was much more jerky than the SGB2's. But this could be a result of my capture card, which can record a maximum of 60.000060 Hz (after separate fields).
The Super Game Boy 2 has its own clock, a proper 4.194304 MHz. The SGB2's refresh rate should be 59.7275 Hz, but the sketchy sources I have seen suggest it is the SNES's 60.0988 MHz. I guess the theory is that a cartridge can drive the SNES faster than 60.0988 Hz but not slower. You can also mod an SGB with the proper clock crystal to run at the canonical 4.194304 MHz. I would hope it would behave identically in terms of framerate, to the SGB2.
I recall that someone on this board indicated that there may have been some benefit to the SGB's method beside being cheaper to make. If there is a discrepancy between the SGB2's framerate output and the SNES's framerate output, then that would suggest that there must be some frame-skipping going on that would not be present with an unmodded SGB.
Since I now own an SGB and an SGB2, I took video captures of a game scrolling left and right. The SGB's capture was much more jerky than the SGB2's. But this could be a result of my capture card, which can record a maximum of 60.000060 Hz (after separate fields).