%pl[native] x/y label positions

Both the MDBX and GAS_DYN2 benchmarks ran fine with FTN95 8.81 (I used /opt /64, and also /debug /64; I think it is a waste of time to attempt to use /check or /checkmate with these old codes).

One problem that you may run into is that GAS_DYN2 writes a 180 Mbyte output file. If the working directory (e.g., on a Ramdisk) does not have that much free space, you may get some strange messages (since the code does not do much checking of I/O).

The SDBG64 version that I have is 8.80.

Thanks for checking, Paul and Mecej4.

Tried same things (if SDBG64 crash with codes compiled 'prog.f90 /64 /debug /link' ) on Intel computer with 880 + recent 881 update.

These two Polyhedron codes crash the SDBG64 debugger if just being copied from AMD computer as EXE files but don't if recompile on Intel without /checkmate. With /checkmate they still crash the debugger

My own codes still crash the SDBG64 debugger (any current or older versions of it) on both 880 and 881 update with any combination of keys. For debugging I can only use pre-880 version of compiler

Mecej4: still with /checkmate do Polyhedron codes crash the debugger or not ?

ALL. If you never tried this great thing like debugger - the way to use it just like this :

SDBG YOURPROG.EXE

Of course you have to compile with /DEBUG and(or) /CHECKMATE. Debugger is simple like AK47, every child can use it 😃 (mine was using it from 2 years old, at least in games like DOOM2 where he was on 77 level at this age. AK47 i mean)

Dan. here is what I find. I compile and link using:

T:\PolyBM>ftn95 /64 /checkmate mdbx.f90 /link
[FTN95/x64 Ver. 8.81.0 Copyright (c) Silverfrost Ltd 1993-2021]

If I run the resulting EXE at the command line, it runs fine. If I attempt to run the same with the SDBG64 debugger, I get:

Error: Access Violation writing address 0x00000000001B1000

000000000040127F

This is on a PC with an i7-10710U CPU, with FTN95 8.81, SDBG64 8.80, Windows 10 Home x64, 32 GB RAM.