Hello Mike - thanks - I clicked the "Doug" link - nothing on runtime vs IDE performance.

Regarding the "UPDATE.EXE" installer - when I test I just copy the EXE to the working folder in "Program Files". Version updates install to a new folder and allow users to import data from an "earlier" version. I don't have to notify and offer patch updates like I used to. The system is fairly stable and my users (bless `em) are much badder a*ses than me - so I try real hard not to embarrass ,yself and look like an idiot - if my users ever learned how much of an idiot I was they'd paint little red dots all over my form and start pulling triggers - my users are derivatives traders - and they take no prisoners and ain't got time to "test" Programming 101 projects.

THey're as demand as most of us here at the UT if that serves as a better measure of my customer base.

It's my first "necessary" try catch And all try catch does is shunt to a less demanding derivative when the math processor attempts to clock above 100% (these are demanding tedious boring random number generators. We all pass variables to functions - but these guys pass functions to function that pass functions to other functions. For your pleasure and consideration consider the following VFP function call:

nBSACA2002 = alfa2 * S ^ Beta - alfa2 * phi(S, t1, Beta, I2, I2, r, b, v) ;
                  + phi(S, t1, 1, I2, I2, r, b, v) - phi(S, t1, 1, I1, I2, r, b, v) ;
                  - X * phi(S, t1, 0, I2, I2, r, b, v) + X * phi(S, t1, 0, I1, I2, r, b, v) ;
                  + alfa1 * phi(S, t1, Beta, I1, I2, r, b, v) - alfa1 * ksi(S, T, Beta, I1, I2, I1, t1, r, b, v) ;
                  + ksi(S, T, 1, I1, I2, I1, t1, r, b, v) - ksi(S, T, 1, X, I2, I1, t1, r, b, v) ;
                  - X * ksi(S, T, 0, I1, I2, I1, t1, r, b, v) + X * ksi(S, T, 0, X, I2, I1, t1, r, b, v)

And that's the simple stuff. I learned that the MSI install (because it implemented the C7 runtime DLL) eliminated most of the Error 78 ("** or ^ domain error.") issues. It's when I start hacking through and doing workarounds for a so called "implied volatility" that I try catch was implemented to shunt to a simpler, perhaps less precise, derivative process.

You'd think the opposite - right - that the MSI runtime would execute algorithms (like the one above) much faster than an IDE runtime would - but - in fact the opposite is true.

So other than buying something from the "Doug", where are his white papers parked in regards to the performance issues under IDE verses MSI runtimes?

And - what is the significance in naming ab MSA UPDATE.EXE (understanding I only issue version upgrades rather than so called "repair" updates"?

TT
www.r6solutions.com

>Do you name the installer update.exe? Is this a vista machine? Doug Hennig wrote quite a bit about installations.
>
>>I suspect it's because the new EXE added a PRG and modified a the structure of a one record DBF.
>>
>>Almost ready to do a new MSI. I am thinking the registry has some kind the map for the (original) installed MSI EXE, and therefore does a lot of "searching" to make sense out the EXE that overlaid it.
>>
>>>>I am Terry's Carbon Unit
>>>
>>>LOL. Please keep references to your unit to yourself out of this family oriented show. ;) If you meant some obscure Star Trek reference, we are all Carbon Units. Lt. Ilia carbon unit was replaced by V'Ger recreation - which was no longer a carbon unit and referred to life forms as carbon units infesting U.S.S. Enterprise, and by extension, Earth and other planets.
>>>
>>>I do not use procedure libraries. I put each function as its own .PRG. These are found directly by the exe. There was a time that issuing SET PROCEDURE TO was a performance hit. That has been reduced.
>>>
>>>I have not seen any performance hits caused by replacing old exe with newer one with my setups. I'm using InnoSetup.
>>>
>>>
>>>>
>>>>LOL
>>>>
>>>>>>The IDE FXP crunch performing better than the EXE crunch is unexpected - seen it before - it went away - the cure may be to simple rebuild the MSI and do a proper install.If it was an issue of doubling up or multiple passes then it seems that would present during the IDE FXP tests.
>>>>>>
>>>>>>Any thoughts appreciated.
>>>>>>
>>>>>>It's not
>>>>>
>>>>>Who are you and what have you done to Terry?
>>>>>
>>>>>Real Terry would have at least 35 spelling errors in a text of this size. You made just this token error, but that is not a good disguise.