I want to mill steel with this machine.

even with that video if you pay attention at one point the tip of the cutter is red hot and tossing sparks. the video is not straight through so in those breaks i know the tool is being swapped out for a new one before the finish pass is done and possibly during roughing as well.

when you get into high speed machining your tools and holders have to be balanced to run at those speeds. they do often run with just air in these applications because coolant would trow the tool into thermal shock. in the video with a VMC those DOC. are really light and they don't look like more than .005 to .010 deep (imperial)

i still think spindle speed or lack of lower ranges of speed is the thing that will hinder cutting steel the most. feed mills work well with harder material with the right inserts but i doubt spending $200 on a holder and another $100 on a box of inserts is that attractive to most. the down side to those kinds of cutters are the are not flat bottom cutters so they don't get all the way through the part unless you have the clearance to do so in the set up.

because the edges are plasma cut most likely that will need slower speeds as well just because of that part of the plan.for the whole plan to be cost effective the tooling needs to last somewhat. with carbide if the rigidity is not there in the machine you will find out fast enough. carbide is great but it loves rigidity. so if the rigidity is not there tooling will chip or break pretty easily.

i have worked with all types of burn outs over the years they are tough no matter what tooling or machine you use. so being able to adjust things to get the job done is important. when you loose the ability to slow things down speed wise the task is even harder yet.

this is all just my 2 cents since this is what i deal with in my trade.

All the tools I have are carbide. I don't have a single HSS tool laying around.

I am working in solidcam so yes HSM machining can be choosen.
I have coolingwater on the machine aswell. I looked at this video, but that is insanely fast, that could never go well with my machine. + The smoothness of motion I could also never reach in linuxcnc , it will give jerks or points where it slows. I posted a topic on this aswell a week or 2 ago and the conclusion was that linuxcnc can't do better because of lacking S curve acceleration and things like that. I had the issue that doing helix entries after every full circle it kind of halts/jerks before moving to the next circle with settings g64 p0.1 q0.1 which is my standard setting through all gcode.

Is it not so that the tool would simply break before anything else would happen to the machine or motors/carriages/guide rails.

I am used to always run above 12000rpm on the spindle. I don't really know how much power is in the lower rpm regions. Never needed to as it's always carbide tools. I know drilling has never been a problem at low feeds only drilling on my machine vises has been an impossible talk as that is a special hardened outer rim which is the hell to go through. But that's only downward action.

I use gwizard for feeds and speed in aluminium and solidcam 2d machining which complete chooses all settings for the cut after choosing what material I am working in. Only used it for aluminium and works like a charm.

On the other hand with a 160a power on the plasma on 12mm mild steel. How accurate would it be possible to make 16mm holes?

carbide has different grades the lower end carbide i think has a hardness of around 65 hrc and the higher end carbide has a hardness around 85 hrc. coatings on the tools can help with tougher material too.

i think with material 12 mm think you might get a hole kinda round but tapper is going to be hard to avoid with the hole sizes no matter what you do on thicker material. plasma is not exactly a precision process in my opinion.

12,000 rpm is pretty fast for steel. if you can get down to 8,000 or less you might have a chance but that's still pretty fast and you are going to be limited to small diameter tools at the speed.

@andy, yes I actually could take out the mill and I have a magcutter here. Have to make something to fit it but that's not a issue.

Is that dirll would be within 0.1 that would do the work indeed.

Still have to do the outside profile of the plate but if I would run that at 0.1mm per pass. It's only 12mm total, that wouldn't be a too lengthy job. I could use a 6mm 4flange cutter running it with coolant

That seems indeed a good idea, as long as this angular cuttur ( I believe as to what you mean is a hollow kind of mill/drill.) Is accurate enough

Do you have anyone laser cutting close to you?

I have but they are just asking so much money for any work.