Re^6: unpacking 6-bit values

That's a neat optimisation, but it gets its gains by pushing some of the required processing out of the benchmark.

What I mean by that is that the sets of 32 numbers are manipulated in pairs (of sets).

So, using uicb() I would have to expand one set to the buffer; copy them somewhere else; then expand the second set; before I could then do the manipulations. Ie, the copying would still need to be done, but it is no longer being measured.

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

"Science is about questioning the status quo. Questioning authority".

In the absence of evidence, opinion is indistinguishable from prejudice.

Comment on Re^6: unpacking 6-bit values

Replies are listed 'Best First'.
Re^7: unpacking 6-bit values by Tux (Canon) on Dec 12, 2010 at 20:46 UTC
My goal here was just to check the influence of pre-bound variables, a-la `bind_columns ()` in DBI and Text::CSV_XS. It wouldn't be too hard to rewrite the init routine to pass it any for of references and make `uicb ()` (Unpack Inline C Bind), store the results in exactly the variables you want and thus bypass all the unwanted copying and the mortalizing cost. The fact that you are dealing with a very limited range makes this very feasible. Enjoy, Have FUN! H.Merijn	[reply] [d/l] [select]

Replies are listed 'Best First'.

Re^7: unpacking 6-bit values
by Tux (Canon) on Dec 12, 2010 at 20:46 UTC

My goal here was just to check the influence of pre-bound variables, a-la bind_columns () in DBI and Text::CSV_XS. It wouldn't be too hard to rewrite the init routine to pass it any for of references and make uicb () (Unpack Inline C Bind), store the results in exactly the variables you want and thus bypass all the unwanted copying and the mortalizing cost.

The fact that you are dealing with a very limited range makes this very feasible.

Enjoy, Have FUN! H.Merijn

[reply]
[d/l]
[select]