Intel® Fortran Compiler
Build applications that can scale for the future with optimized code designed for Intel® Xeon® and compatible processors.

Haar Function

New Contributor I

I dragged out some very old code to do a Haar Wavelet.  

The standard input is taken from Unit 5 and Unit 6.  I had completely forgotten those unit numbers use.

INTEGER (KIND=4), PARAMETER                :: INPUT_UNIT                   = 5


This is from the iso_fortran module, should I use the module or *. Which is preferred? 

0 Kudos
7 Replies
New Contributor I

Dear Intel Guru:

many years ago, when I used to have to pay for Intel Fortran and had license subscriptions,  I ended up with all of the licenses attached to  I have had this for so long and Opera automatically remembered the password, but I recently reinstalled and can not find my copy of the password.  Interestingly it is more than a decade since I used the gmail account and so I cannot recover it, again forgotten password. 

Is there anyway I can reset the password and continue to use the gmail account at Intel, it is not big deal, I am just enjoying getting the notes that I am a beginner.  



Black Belt Retired Employee

Regarding READ, feel free to use *. The main use of xxx_UNIT definitions is if you want to change properties using OPEN (see Doctor Fortran in "The Modes, They are A-Changin'" - Doctor Fortran (


As for your license issues, ask in Registration, Download, Licensing and Installation - Intel Community

Black Belt

"The standard input is taken from Unit 5 and Unit 6."

That is probably not quite correct. Unit 6 was routinely used for output, not input.

New Contributor I

Yes, you are correct, I made a mistake. 

New Contributor I

Vehicle on BridgeVehicle on Bridge

I was looking for the HAAR function to pull out a vehicle load from the thermal stream.  As you develop software, the interesting problem is people ask, can it do this?  In this case, the this is, can you determine the bridge damping, of course in real time and is it statistically changing.  

You cannot determine the bridge damping from the displacement graphs, a vehicle travelling over a bridge creates interesting displacement profiles, that fit the Czech mathematics, but are lousy and slow for damping analysis. The problem is a car is not a nice hammer on a beam as is traditionally shown in the papers.  

The picture shows reality as against the nice smooth sine wave from the papers.  So starting with the primary decay function of long duration, the trick is to develop some Fortran code that picks out the top 10 peaks on the upside and the downside, given that the data is in a 1 D vector.  

I was thinking to reverse the vector,  and do some form of simple hill climb function, only pick up the peaks

So to start with I added DISLIN.  As always interesting, you have to find a working DISLIN with and old VS project and then make sure you copy the exact settings,  and there are a lot of them now.  

1>LINK : warning LNK4098: defaultlib 'LIBCMT' conflicts with use of other libs; use /NODEFAULTLIB:library

I get this warning, in DISLIN you now need to use the legacy*.lib, gdi32 and user32 libraries.  Can I fix this or jus live with the comment? 




Black Belt

The above chart (visually) shows two damping's one at 400/3 (for large displacements) and one at about 400/4.75 or so.


Black Belt Retired Employee

If you're building DISLIN, make sure its runtime library settings are the same as for your executable project (static vs. DLL, debug vs. non-debug). In MSVC, these are under "code generation".