forums.silverfrost.com

[silverfrost]

There is a Wikipedia entry for FTN95 that I have edited and updated over the years. Wikipedia have scheduled it for deletion which seems a shame as there are now very few Fortran compilers around. If you have anything to contribute to the page (anyone can edit Wikipedia articles) then please do.

https://en.wikipedia.org/wiki/Silverfrost_FTN95

[mecej4]

I was not aware of the existence of the Wikipedia article on FTN95, even though I frequently use Wikipedia. Perhaps, the article was born after I had already been using this Forum and FTN95 for many years.

I did have a long-unused account on Wikipedia, and I was able to log in.

I added a paragraph on SDBG/SDBG64 and edited out the "marked for deletion" notice. It is many years since I made edits on Wikipedia, so I have no inkling as to whether my editing is enough to keep the page alive.

[LitusSaxonicum]

Mecej4,

Is it worth noting that the compiler has an option to produce 64-bit code? And an integrated development environment, PLATO? And a standalone resource compiler / linker / librarian ? Online help files as well as inbuilt help for PLATO ?

Eddie

[mecej4]

Eddie,

The Wikipedia article does cover the attributes that you asked about (Robert probably was the author). I added a paragraph about SDBG as a way of refreshing the page and stalling the sunset action.

You can add content or edit the wording on any Wikipedia article, but note that someone else can then change what you wrote.

I don't think that Wikipedia needs to reproduce what is in the Silverfrost pages. If the Wikipedia article on FTN95 helps direct some traffic to the Silverfrost web pages, it should suffice.

I have mixed feelings about scientific content on Wikipedia. Some articles are very good. Some others cover too much in one arcane subtopic, and a few articles are trash. Very different from, say, the 1911 Encyclopaedia Britannica.

[simon]

The page says that 8.95 has been available since January. The latest updates page https://forums.silverfrost.com/viewtopic.php?t=4245 is only at 8.92, dated October. Have I missed an update?

[mecej4]

There is usually a gap of a couple of months between the release of a new version to registered customers and the release of the same version for personal use (without registration).

I checked just now, and I see that the 8.95 version is now available for personal use.

[simon]

Ah yes. Thanks. I should have looked at the announcements pages rather than the latest updates page (i.e., https://forums.silverfrost.com/viewtopic.php?t=4245). It would be good to get those synchronised.

[DanRRight]

Need some healthy refreshing critics? Sometimes we all do. And here do not watch anywhere besides me . Make some correction though on my poor English as it might hurt even more.

This page looks more like an anti-ad for me after comparison with similar Intel Fortran one. The most of negativity comes from the first obvious conflict thought why in 2023 users need 1995? Everyone knows that in older years compilers were often called to reflect the Standard's year like Fortran66 or Fortran77 for 1966 and 1977. Hence at least mentioning that the work to incorporate the latest Fortran Standards to be modern and relevant to present day and future trends is permanently takes place.

Another negative moment is the lack of visuals. This unfortunately is common for all Fortran compilers. The text description can impress only English natives, where the fraction of fluent readers also nosediving to zero due to latest decades methods of education. I am the last non-native English reader who reads some English texts sometimes, and even i almost never RTFM . Show people pictures, examples, videos, blogs, and even in the error report - the places where errors were found, instead of writing and describing that by words. I value Salford Examples which are now 25 years old as the best what company have done in its advertisement. I would never look at Clearwin, graphics library, bitmaps handling, animations, OpenGL and multitasking (starting with unseen anywhere else multitasking under DOS in pre-Windows era) without them and by now Intel Fortran would have one more customer.

Some good points of ads from previous Salford compilers were also not stressed out. But their goals indeed were achieved: with this compiler you get universality when science and engineering programmer mostly do not need to look anywhere else and if this becomes needed this could be also achieved

Not mentioned were geekish things like native support of HTML if you need to make your code Help description or native Assembler support straight in the mix with the Fortran code for ultimate nuts. OpenGL was mentioned but all others do that too, while this compiler supports OpenGL natively written in plain Fortran. Text and graphics handling on graphics screen is also the powerful feature no others have. This allows creating versatile plots with professional quality with just one line of Fortran code and easily adjust anything you want afterwards without any programming just by clicking on the buttons and menus.

By the way mentioned Simfit needs to remove and renew 3D, Cylinder and Pie chart plots as utterly obsolete anti-ad examples causing sickness in young generation for its low quality. https://simfit.silverfrost.com/gallery.html No employer would hire them if they show such job quality in their application. I do not know how many times i already wrote that and even posted good examples. As a minimum Simfit have to stop using CGA 16 color palette. And redo all bar charts, they look ugly.

Missing the sentence why VS2022 is not supported. Is this because of too many issues across many other compilers? Then mention that not currently supported but work is in progress

Linux version absence probably have to be commented too if you at least evaluate this possibility

[Kenneth_Smith]

This would be a good example to show under Clearwin+ along with the image it produces (I don't know enough about loading images to Wikipedia to do this myself).

[DanRRight]

Ken is a tricks hack, his approaches turn me upside down.

I am kind of style hack. My other goals are ultimate simplicity and maximum speed.
No, here is what i am talking about: just ***one*** line of Fortran code and you get profy quality plot not achievable with any other approach on earth. Here is entire code. You compile it and reuse for other data and cases. When you need to change anything there is no need to recompile, resize it, or just click on the icon and change the color of plot, line style, text size and position, anything

[DanRRight]

And here is another example of same sort. I think it's classics already and deserves not just the Wiki but probably Guinness record book. What one line of Fortran code can do in Clearwin:

[Kenneth_Smith]

[mecej4]

Kenneth's comments reminded me of:

[Kenneth_Smith]

I have added a small example making use of %pl. Can be be removed if not appropriate. No programming tricks to rotate Dan or others.

[DanRRight]

Will this convince MATLAB users to switch? Look how easy and shiny same is done there:
https://www.mathworks.com/matlabcentral/answers/793702-how-do-i-get-a-graph-to-spiral
https://www.mathworks.com/matlabcentral/answers/815085-creating-a-3d-helix-spiral-with-surf-command