Lines Matching refs:gain

262 No pain. no gain.
474 Unrolling twice keeps the gain at about 15%
475 Unrolling three times gives a gain of 30% compared to attempt 2.
480 feeling that in the next steps I would obtain additional gain from it.
559 Measuring this code again showed big gain. When executing the original
658 shift after parity[rp7] << 7; No gain.
665 No gain.
690 It proved to be fastest to do some table lookups. Performance gain
700 The gain when calculating the ecc is tremendous. Om my development hardware
705 For correction not much gain could be obtained (as bitflips are rare). Then
708 It seems there is not much more gain possible in this, at least when