Anand's Mobility Radeon 9600 vs GeforceFX go5650 review. . .
- Thread starter Ostsol
- Start date
Sorry, I haven't read this thread through, but just wanted to pick up on a point. Sorry if it was already addressed.
I made this point in another thread recently, but it pertinent here as well. I've wrestled with the idea of "like for like" or "how the user will play it" - with the TR:AOD comparison benchmarks I actually ran both because I figured there would be complaints if saying "its too slow, you'd never play it like that", so not only did I use our "apples to apples" settings but also the default game settings for each board and listing what is enabled in these modes.
However, in reality this is this is not the issue for a hardware review. A hardware review is not about justifying the performance for current users, its about informing potential users on the best purchase. If there is a board that can run the full settings at acceptible levels, but others are slower (to the point of non-playability) surely that tells the potential user which would be the most sensible purchase for that particular title. If you are going out an buying a board are you really that concerned that board X running in DX8 is just about as fast as board Y running full DX9 settings? If overall performance is an issue, surely the board that performs well at full detail settings also has the greatest room to increase performance because there is more room to turn more features off/down.
Ostsol said:
I made this point in another thread recently, but it pertinent here as well. I've wrestled with the idea of "like for like" or "how the user will play it" - with the TR:AOD comparison benchmarks I actually ran both because I figured there would be complaints if saying "its too slow, you'd never play it like that", so not only did I use our "apples to apples" settings but also the default game settings for each board and listing what is enabled in these modes.
However, in reality this is this is not the issue for a hardware review. A hardware review is not about justifying the performance for current users, its about informing potential users on the best purchase. If there is a board that can run the full settings at acceptible levels, but others are slower (to the point of non-playability) surely that tells the potential user which would be the most sensible purchase for that particular title. If you are going out an buying a board are you really that concerned that board X running in DX8 is just about as fast as board Y running full DX9 settings? If overall performance is an issue, surely the board that performs well at full detail settings also has the greatest room to increase performance because there is more room to turn more features off/down.