| 1 |
On 11/07/2014 07:54 PM, Matthias Maier wrote: |
| 2 |
>> Well, you're not comparing like with like. Paludis with "everything |
| 3 |
>> turned off" does more than Portage with "everything turned on". If all |
| 4 |
>> you're looking for is the wrong answer as fast as possible, there are |
| 5 |
>> easier ways of getting it... |
| 6 |
> |
| 7 |
> The last time I compared the resolver speed of portage and paludis both |
| 8 |
> needed almost the same time. |
| 9 |
> |
| 10 |
> Do you have a speed comparison with a similar feature set of both? (Or, |
| 11 |
> alternatively, the speedup one gains by tuning paludis to be as fast as |
| 12 |
> possible). |
| 13 |
> |
| 14 |
|
| 15 |
I think you didn't get the idea: it doesn't make much sense to compare |
| 16 |
the speed if the correctness differs. |
| 17 |
|
| 18 |
Also, I don't understand these discussions. The time dependency |
| 19 |
resolving takes is marginal compared to the whole update process, no |
| 20 |
matter what PM you use. |
Archives