instead of increasing core clock for Linux - reduce it for Windows
This commit is contained in:
parent
c96851dbc3
commit
ebf788e916
@ -198,7 +198,6 @@ func readMangoHudFile(scanner *bufio.Scanner) (*BenchmarkData, error) {
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("failed to parse GPU mem clock value '%s': %v", record[7], err)
|
||||
}
|
||||
val = math.Round(val*2*100000) / 100000 // multiply by 2 (so it represents actual core clocks) and round to 5 decimal places
|
||||
benchmarkData.DataGPUMemClock = append(benchmarkData.DataGPUMemClock, val)
|
||||
|
||||
val, err = strconv.ParseFloat(record[8], 64)
|
||||
@ -374,6 +373,7 @@ func readAfterburnerFile(scanner *bufio.Scanner) (*BenchmarkData, error) {
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("failed to parse Memory clock value '%s': %v", record[7], err)
|
||||
}
|
||||
val = math.Round(val/2*100000) / 100000 // divide by 2 (so it represents core clocks in a same manner as Linux) and round to 5 decimal places
|
||||
benchmarkData.DataGPUMemClock = append(benchmarkData.DataGPUMemClock, val)
|
||||
}
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user