Dropping Array Visualizer from IVF and IVC

Dropping Array Visualizer from IVF and IVC

I was very suprised to be notified of Intel's decision to drop AV from IV FORTRAN and IV C++. This was one of the best feature reasons for purchasing the Intel compilers.

Jim Dempsey

Notice received today:

Intel will not be including the Intel Array Visualizer component in the next release of Intel C++ and Fortran Compilers for Windows*. After the next release of Intel Compilers, the most recent version of the Intel Array Visualizer will be made available as a free, unsupported download from <http://www.intel.com/cd/software/products/asmo-na/eng/compilers/226277.htm> and will no longer require an Intel Compiler license to use. Intel will no longer maintain or develop the Intel Array Visualizer. The Intel Array Visualizer user forum will continue to be available at <http://softwareforums.intel.com/ids/board?board.id=Visualizer> where users can share experiences and usage tips. Please visit the forum for more information.

10 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

I totally concur.
Dropping Array Visualizer and not spending any more money to develop it is disastrous. I guess it has something to do with John Readey who is the chief architect of that. He hasnt responded lately to any messages in the forum. He has done an excellent job.
I am willing to pay extra for this product.

I have current commercial applications using this product and need to support it in future.


Yes, John, "Mr. Array Visualizer", chose to move on from Intel. We all liked John and regret his leaving - I've worked with John for nearly ten years - but we wish him well in his new endeavours. Unfortunately, AV was pretty much an "all John" project and we just don't have anyone else who can fill his shoes.

Bonnie Aona, of our support team, will participate in this forum. She had been getting training from John on use of AV and she may be able to answer some questions. She'll introduce herself here shortly. We hope that other AV users will also share their knowledge.

AV is not "going away" - the latest version we have will be made available as a free, unsupported download, so your applications will still work. We won't be further developing or maintaining it, however.

This was a hard decision to make. We know that AV is very popular with its users, but we just didn't see a way to do it justice going forward.

Retired 12/31/2016

Array Visualizer is/was a very important part of IVF and I decided to migrate to this compiler because of its existence. Furthermore I started to migrate the visualization part of my software to AV. Now I received a message that it was one-man project and nobody is going to take it over. No further development and patches? It is not fair and it is non professional attitude.


I think part of the problem (if there were a problem) was a lack of understanding by marketing that the Array Visualizer could be used as one of the strongest selling points for Intel Visual Fortran. Visual Fortran is not nearly as Visual without this product. In my opinion, marketing, not having a programming background, and more importantly not using the Array Visualizer, had absolutely no comprehension of what could be done with the Array Visualizer. The core of this problem is the product is misnamed. By use of name alone, Array Visualizer would seem to be some sort of means to dump array data in a visual way. This is an understatement of the full capability of the product.

I strongly suggest that before this golden nugget slips out of their fingers that the V.P. of marketing takes the time to visit some of his users that incorporate the Array Visualizer into their software in a significant manner. Only by doing so will he/she grasp the value of this feature of Intel Visual Fortran.

If the Array Visualizer is the product of one person (John Readey) then I would suggest that Intel entice him back with whatever it takes (bonus, salary, funding, respect, etc...). Virtually everything is present for Intel Marketing to exploit the Array Visualizer into sales for Intel Visual Fortran. A little more work on documentation and a hefty amount of promotion is all that is needed to make a winner out of this combination.

Take my project for example. I am running simulations of a second generation space elevator. This is a dynamic (moving) tension structure attached to Earths equator and stretching out about 70,000 miles. This structure consists of 13 objects each with state history, 18 tethers each with 1000s of data points and each with 7 charts and then a rather crude quad-mesh of Earth. The entire graph is animated to run during the simulation of the space elevator. While running I can pan, rotate and zoom in/out to observe events in the simulation as they occur.

The integration of a "live" graphical representation of what is happening in the simulation is an invaluable tool for my research. It saves me time, gives me a visual feedback to what is happening, and makes a gee-whiz demonstration. This is the motivating reason for me to buy Intel Visual Fortran instead of downloading a "free" copy of an Open Source Fortran (support was another reason).

When I made this purchase decision, and due to lack of exposure by Intel marketing, I had absolutely no clue of what the Array Visualizer could do for me. I thought "it might be useful". As it turned out, after managing a rather steep learning curve (fault of documentation), I find it indispensable to my research requirements.

Although most of the Array Visualizer is Open Source, the integration into Intel Visual Fortran (and Windows) is what counts. I am certain other users of AV find it equally important to them as I have found it for me.


Jim Dempsey

The Intel Array Visualizer was a promising feature, but I found it often frustrating. I wrote a module that encapsulated some of the more painful
aspects of dealing with AV, such as autoscaling axes for dynamic plots.

- What reaction would Intel have if AV was put on sourceforge?
- Would anybody be willing to contribute time/code to its development?
- Is plplot, etc or OpenDX a better solution?
- Would Intel GPL Array Visualizer?

We are open to the idea of open-sourcing AV, but there are a number of issues to deal with in order to do that. Just "throwing it over the wall" is not in itself a solution - it would really need someone willing to take responsibility for it.

Retired 12/31/2016

I am also very disappointed by this step of Intel. I have migrated from CVFto IVF but now I see that IVFprobably neverreach level of quality of CVF.

I would like to know if it is possible to use the CVF libraries for the Compaq Array Visualizer in Intel Visual Fortran?



We worked closely with John and made extensive efforts to leverage the Array Viewer capabilities to provide enhanced data visualization for our analysis package O-Matrix, http://www.omatrix.com/overview.html. We created a set of O-Matrix interface functions that provide O-Matrix language control of the Array Viewer. Please see our add-on page at: http://www.omatrix.com/datavis.html for details of this library.

At this point a significant percentage of our users rely on this library to provide additional plotting capabilities from O-Matrix. Intel's failure to provide continued support for AV will be a large distress for our users if we also have to subsequently drop support for AV from within O-Matrix. We highly encourage Intel to re-consider their support position or to make the entire AV source code available within the public domain s.t. it can be maintaine and enhanced. If the source code base is made available I highly suspect that our company would be willing to take an active role in AV maintenance, support, and enhancements.

Beau Paisley
Harmonic Software Inc., www.omatrix.com

Leave a Comment

Please sign in to add a comment. Not a member? Join today