Jump to content

  • Log In with Google      Sign In   
  • Create Account

Banner advertising on our site currently available from just $5!


1. Learn about the promo. 2. Sign up for GDNet+. 3. Set up your advert!


#Actualpatishi

Posted 16 July 2013 - 06:16 AM

I am trying to implement the "Deep + always"  or  the "Two tier System"  .. e.g   at every index in the table, i have one entry that gets replaced by "depth" , that means that only if the depth is higher (i assume that means closer to the root node) it gets replaced by the new entry.      And the second entry always gets replaced no matter what.   

The question i have regarding this:     when i want to insert new entry to the table,  than i go to that specific index and first check the "by depth" entry,  and if the depth is higher than the one in the table, i immediately switch between them, i even don't check if this is the same position or just another one hashed to the same index.   
But if not (!)  ,e.g, the depth is not higher, than i need now to put the new entry in the second poisition (the "always replace "one).     but i didn't get if i need to check whether the new position is different than the one i already have in the first position (the "depth " entry)  or i allowed to have two identical entries, side by side.        

my logic tells me that it is a bit of waste to have two identical positions one next to the other, when in any case,  when probing for the table for an entry, i always check the "depth" entry first anyway...     I hope that the question is clear.smile.png       in other words, do i need to check if the two entries are similar or not.
 


#2patishi

Posted 16 July 2013 - 06:15 AM

I am trying to implement the "Deep + always"  or  the "Two tier System"  .. e.g   at every index in the table, i have one entry that gets replaced by "depth" , that means that only if the depth is higher (i assume that means closer to the root node) it gets replaced by the new entry.      And the second entry always gets replaced no matter what.   

The question i have regarding this:     when i want to insert new entry to the table,  than i go to that specific index and first check the "by depth" entry,  and if the depth is higher than the one in the table, i immediately switch between them, i even don't check if this is the same position or just another one hashed to the same index.   
But if not (!)  ,e.g, the depth is not higher, than i need now to put the new entry in the second poisition (the "always replace "one).     but i didn't get if i need to check whether the new position is different than the one i already have in the first position (the "depth " entry)  or i allowed to have two identical entries, side by side.        

my logic tells me that it is a bit od waste to have two identical positions one next to the other, when in any case,  when probing for the table for an entry, i always check the "depth" entry first anyway...     I hope that the question is clear.:)       in other words, do i need to check if the two entries are similar or not.
 


#1patishi

Posted 16 July 2013 - 06:14 AM

I am trying to implement the "Deep + always"  or  the "Two tier System"  .. e.g   at every index in the table, i have one entry that gets replaced by "depth" , that means that only if the depth is higher (i assume that means closer to the root node) it gets replaced by the new entry.      And the second entry always gets replaced no matter what.   

The question i have regarding this:     when i want to insert new entry to the table,  than i go to that specific index and first check the "by depth" entry,  and if the depth is higher than the one in the table, i immediately switch between them, i even don't check if this is the same position or just another one hashed to the same index.   
But if not (!)  ,e.g, the depth is not higher, than i need now to put the new entry in the second poisition (the "always replace "one).     but i didn't get if i need to check whether the new position is different than the one i already have in the first position (the "depth " entry)  or i allowed to have two identical entries, side by side.        

my logic tells me that it is a bit od waste to have two identical positions one next to the other, when in any case,  when probing for the table for an entry, i always check the "depth" entry first anyway...
 


PARTNERS