"After conversions to 8.0, I had a few instances of reports that were slower than they had been in 7.6."
I would imagine that this was done on the same machine with the same hardware....?..... "I have a theory -- based on little knowledge -- that there are some formulas in the "query optimizer" that do some hard-coded math with assumptions about 32-bit pointers in indexes. In 8.0, the pointers are 64-bits." If that is the case: would utilizing a 64 bit OS make any difference? Chr*s ________________________________ From: [email protected] [mailto:[email protected]] On Behalf Of Bill Downall Sent: Thursday, May 07, 2009 8:06 AM To: RBASE-L Mailing List Subject: [RBASE-L] - Re: 7.5 vs V8 speed After conversions to 8.0, I had a few instances of reports that were slower than they had been in 7.6. But in my case, the project-to-temp-table solution alternative made them much faster. I have a theory -- based on little knowledge -- that there are some formulas in the "query optimizer" that do some hard-coded math with assumptions about 32-bit pointers in indexes. In 8.0, the pointers are 64-bits. My only evidence to support this theory is that the "same" query optimizer and the "same" indexes give different results. What has changed? Bill On Thu, May 7, 2009 at 8:39 AM, Lawrence Lustig <[email protected]> wrote: << Upon reconsideration, you're probably right, Larry. I should keep my brain away from the keyboard after a certain hour. >> No worries. I have to say that I'm intrigued by this issue that Marc is having. Although I have a copy of 8.0 we don't yet have any production applications that use it. I'd hate to think that anything actually gets _slower_ moving to 8.0, so I'm hoping Marc posts the eventual solution to his problem. -- Larry
