Here's the difference between natural flint and ferrocerrium flints...
When one strikes a Ferro flint, the ferro material comes off and creates a spark. When one strikes a Natural flint, the steel comes off and creates a spark. For that, you want a soft steel. You don't want anything tempered. I hear that iron works very well.
The theory is that the friction of the strike causes a droplet of red-hot steel to be struck off the steel. Inspection under magnification shows droplets of steel produced by the strike. "Too hard" defeats the effort. "Too soft" means insufficient friction; you may simply shave off steel without producing maximum heat.
As for practice:
I would defer to Ragnar. His steels work.
We have a maker of fire steels here in Ohio who tempers and heat-treats his steels. I have seen them made. They work. He says they are "knife hard," whatever that means.
I have used old files with great success and seen them used many times with success. They are hardly dead soft. Pretty damn hard in fact.
Musket lock steels were not dead soft.
I have had success striking sparks with flint and a Western sheath knife and a Cattaragus 225Q. How hard? Unknown but clearly not dead soft.
I suggest that theory and experience dictates that dead soft steel makes a relatively poor fire steel.
Well I spoke to Ragnar via email and this is what I learned from him. First off you want hard steel rather than soft steel (this completely surprised me). With soft steel you take off too much. He recommended RC60 as a minimum (if I am remembering the email correctly). He also suggested that I round the back of one of my frost Moras with a file to get past the lamination to the core steel which he sais is around RC60-61. I did this and it worked like a charm! Not to mention that modifying knives is always fun too! So to those who suggested I contact him, you were correct. He responded very quickly and he was very helpful and polite. I'm looking forward to sending more of my money his way. Great guy!