Support Forums - View Thread

View Thread

The following is the text of the current message along with any replies.

Messages 1 to 10 of 10 total

Sat, May 9 2009 11:28 PM	Permanent Link
"Jerry Clancy"	D5 already has enough problems with debugging in XP, but I found yet another one. After updating DBISAM to Build 5 I decided to try FastMM. After sorting out the issue it has with madExcept I finally got it working but found that one could not restart from a breakpoint with it installed. Have no idea why so I just removed it from the .dpr file and the problem went away. There were no apparent problems at deployment, though, at least so far. Jerry
Mon, May 11 2009 5:03 PM	Permanent Link
Tim Young [Elevate Software] Elevate Software, Inc. timyoung@elevatesoft.com	Jerry, << D5 already has enough problems with debugging in XP, but I found yet another one. After updating DBISAM to Build 5 I decided to try FastMM. After sorting out the issue it has with madExcept I finally got it working but found that one could not restart from a breakpoint with it installed. Have no idea why so I just removed it from the .dpr file and the problem went away. There were no apparent problems at deployment, though, at least so far. >> I haven't seen this one yet with both FastMM and D5. The only problem I'm aware of with D5/XP is the issue with debugger symbols and DLLs. -- Tim Young Elevate Software www.elevatesoft.com
Mon, May 11 2009 11:43 PM	Permanent Link
"Jerry Clancy"	Tim, I've got workarounds for most of those debugger problems, but this was the first time I tried to use FastMM and that was the result. Ripped it out. Jerry "Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message news:12D7284A-7586-4EF1-85F8-AB4AC7C8DC65@news.elevatesoft.com... > Jerry, > > << D5 already has enough problems with debugging in XP, but I found yet > another one. After updating DBISAM to Build 5 I decided to try FastMM. > After sorting out the issue it has with madExcept I finally got it working > but found that one could not restart from a breakpoint with it installed. > Have no idea why so I just removed it from the .dpr file and the problem > went away. There were no apparent problems at deployment, though, at least > so far. >> > > I haven't seen this one yet with both FastMM and D5. The only problem I'm > aware of with D5/XP is the issue with debugger symbols and DLLs. > > -- > Tim Young > Elevate Software > www.elevatesoft.com >
Wed, May 13 2009 2:44 PM	Permanent Link
Tim Young [Elevate Software] Elevate Software, Inc. timyoung@elevatesoft.com	Jerry, << I've got workarounds for most of those debugger problems, but this was the first time I tried to use FastMM and that was the result. Ripped it out. >> Did you try it without MadExcept ? It may be an issue there. We use FastMM4 in all of the .exes that we distribute, including the DBISAM Database Server, and they're all compiled using D5. -- Tim Young Elevate Software www.elevatesoft.com
Wed, May 13 2009 5:57 PM	Permanent Link
"Jerry Clancy"	No, I didn't, though I had to prevent it from modifying the load sequence. I'll try running the .dpr tonight with madExcept disabled. BTW, I know it always depends upon the nature of the app but how significant is the potential performance hit if you run without a memory manager? "Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message news:3DFFB0DC-EB81-408B-86AF-9F15F60C36C5@news.elevatesoft.com... > Jerry, > > Did you try it without MadExcept ? It may be an issue there. We use > FastMM4 in all of the .exes that we distribute, including the DBISAM > Database Server, and they're all compiled using D5.
Thu, May 14 2009 10:38 AM	Permanent Link
"Jerry Clancy"	Tim, Same issue without MadExcept. Jerry "Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message news:3DFFB0DC-EB81-408B-86AF-9F15F60C36C5@news.elevatesoft.com... > Jerry, > > << I've got workarounds for most of those debugger problems, but this was > the first time I tried to use FastMM and that was the result. Ripped it > out. > >> > > Did you try it without MadExcept ? It may be an issue there. We use > FastMM4 in all of the .exes that we distribute, including the DBISAM > Database Server, and they're all compiled using D5. > > -- > Tim Young > Elevate Software > www.elevatesoft.com >
Fri, May 15 2009 11:08 AM	Permanent Link
Tim Young [Elevate Software] Elevate Software, Inc. timyoung@elevatesoft.com	Jerry, << BTW, I know it always depends upon the nature of the app but how significant is the potential performance hit if you run without a memory manager? >> Multi-threaded on a multi-processor or multi-core processor - significant difference. Single-threaded or multi-threaded on a single-processor with a single core - no noticeable difference. -- Tim Young Elevate Software www.elevatesoft.com
Fri, May 15 2009 11:08 AM	Permanent Link
Tim Young [Elevate Software] Elevate Software, Inc. timyoung@elevatesoft.com	Jerry, << Same issue without MadExcept. >> Wow, that's really odd. Are you compiling using the source units of FastMM4 ? -- Tim Young Elevate Software www.elevatesoft.com
Sat, May 16 2009 12:24 AM	Permanent Link
"Jerry Clancy"	Good to know. I'm in the latter group for now. Jerry "Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message news:E4540C74-30EC-4456-98BD-DB644B5B5C0E@news.elevatesoft.com... > Jerry, > > << BTW, I know it always depends upon the nature of the app but how > significant is the potential performance hit if you run without a memory > manager? >> > > Multi-threaded on a multi-processor or multi-core processor - significant > difference. > > Single-threaded or multi-threaded on a single-processor with a single > core - no noticeable difference.
Sat, May 16 2009 12:25 AM	Permanent Link
"Jerry Clancy"	Yes. Used their supplied .pas unit. "Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message news:CA91B429-98D9-43D0-B52C-620D27D95B5A@news.elevatesoft.com... > Jerry, > > << Same issue without MadExcept. >> > > Wow, that's really odd. Are you compiling using the source units of > FastMM4 ?