Associate
- Joined
- 30 Aug 2009
- Posts
- 467
@Warsam71, I know that it'll have been dictated top down, probably after so much negative feedback from users of GCN 1.0 cards, but AMD really should have released the GCN 1.1 version of the Mantle driver and followed up with additional 14.1 beta updates to include GCN 1.0 support.
If it's a show stopper bug with GCN 1.0 then the last thing that should be done is attempt to flog the engineers into working longer and faster to hack the bug out. It's not a long term solution and it usually takes 2.5x as long for the solution to be implemented when people are tired and stressed.
Whoever made the cockup of a decision to try and get Mantle out for all GCN in two days likely didn't consult the actual engineers themselves when they made the call. Engineers will always give three timeframes, at least, for stuff like this. One will be if the stars align and everything goes perfectly, the second is the best guess as the realistic earliest time to expect a resolution and thirdly the absolute longest that it should take.
Some idiots in management jump on the earliest figure and think that they can force that to happen, and if it passes they think that they can jump on the team and flog them harder to get the issue resolved sooner. It almost never happens like that though because stress and tiredness inhibit the the cognitive processes required to get the problem resolved.
If it's a show stopper bug with GCN 1.0 then the last thing that should be done is attempt to flog the engineers into working longer and faster to hack the bug out. It's not a long term solution and it usually takes 2.5x as long for the solution to be implemented when people are tired and stressed.
Whoever made the cockup of a decision to try and get Mantle out for all GCN in two days likely didn't consult the actual engineers themselves when they made the call. Engineers will always give three timeframes, at least, for stuff like this. One will be if the stars align and everything goes perfectly, the second is the best guess as the realistic earliest time to expect a resolution and thirdly the absolute longest that it should take.
Some idiots in management jump on the earliest figure and think that they can force that to happen, and if it passes they think that they can jump on the team and flog them harder to get the issue resolved sooner. It almost never happens like that though because stress and tiredness inhibit the the cognitive processes required to get the problem resolved.