I can only tell you my opinion based on my experience. I would not go for either a belt driven or non proper liniar rail machine when looking for 3d routing or carving materials harder than softer wood. The x-carve has both, and an not very smart construction on top of that. The portal is so soft, it twists visible by just driving the router into it's middle. And a z extention will make this even worse. And just slower isn't always the answer. I can tell you, that the vibrations both from the belts and from the flex will ruin your surface at least in a way, that you should start sanding with 120 grit paper.
And stiffening it... well, the construction is so bad, why should you try to fix it, if a good solution isn't that hard at all. maybe the improvements make it a good machine. But someone who thinks that the x-carve is a solid base to start a machine from with some improvements, has at least a different understanding of a good cnc router than me.
The x-carve is a low budget and router for 2d cutting of soft materials. Even the producer says so. Every construction detail is made with this in mind. from the very weak open loop steppers to the cheap bearings. It is not just the stiffness

Well, I didn't mean to turn this into a discussion on machine suitability, but what is your opinion on the Stepcraft then? I can stretch the budget for a Stepcraft, and knowing that EC are using it for what I want to do with it makes it appealing as well, but I just can't really see much difference between the two from my - admittedly very inexperienced - perspective. I know the Stepcraft will have better quality components, but I'm not sold on whether they are actually necessary. The covered bearings are nice, but it's a half hour job to make something similar for the X-Carve. It's stiffer, but again I'm not opposed to just stiffening the frame of the X-Carve if I find it is loosing steps. 

Realistically, it's 2.5x the price of the X-Carve, and a lot of that will be due to it's flexibility (function wise, not stiffness wise) which I don't need. But is there a reason the Stepcraft can do 3D routing of tooling board, and the X-Carve would never be able to? To me, it looks like the X-Carve can be modified to do what I need it to, comes with a bigger bed, and still comes in at less than half the price of the Stepcraft?

To be fair, I am looking at this as a way to reduce my manual pattern making. Hand carving polyurethane foam, even with the 2D profile skeleton structure, can only be so accurate. So I'm only really looking to replicate those tolerances, I'm not trying to get into precision milling of high performance engine components. 

I would not go as far as to recommend a stepcraft for 3d milling. Compared to the x-carve the stepcraft machines are real rigidity monsters. However, mechanically even the stepcraft are closer to 3d printers than to proper cnc machines. And I am not talking about industrial steel machining. Stepcraft is very expensive for what you get.
Have a look at sorotec.de. starting  with the basic line, this are the kind of machines I would be looking for. Sorotec might not be an option in Australia, but there should be people building cnc machines like that. What you should look for, in my opinion, are proper linear bearings (wich both x-carve and stepcraft are missing) and  ball screws (same here, stepcraft has at least screws, belts are not suited for any cnc milling in my opinion), And a stiff portal (is that the right word in English?) (Here, the smaller stepcrafts might be ok,  but the x-carve is falling way behind). Additionally, usb has got a real bad reputation in the cnc community, because it might cause some problems. I would go for parallel port, if the budged is tight. Also I would like to mention, that a cheap desktop pc will be an advantage when it comes to cnc working. The energy saving systems of laptops can cause some quite strange problems.

Have a look at ec's video again. This is no stepcraft machine. It is hard to tell what kind of machine it is, but it looks like some China machines I saw on eBay. At least it should have unsupported rail bearings, or even supported. The second would would be the lowest standard i would be looking for. And it definitely has a more powerful spindle.
If you got no need for precision, a stepcraft might be ok, but it will take a stepcraft much longer than the machine in the video. If an x-carve could achieve those results, I don't know. I doubt it.

Well spotted in the video, I just took another look and you're quite right, they aren't using the Stepcraft for that tutorial. I had assumed given the other tutorial on CNC machining carbon flat sheet that they were using the same machine in-house, so glad you picked that up! 

Ok, well I shall probably rethink this then and stick to the traditional manual methods. Appreciate the input oekmont!

It just happend, that while we were texting here, I bought an upgrade for my cnc solution. 1000 pounds of stiffness here. This machine can do some serious aluminium milling. I will have to swap the computer however, because a laptop,as I wrote earlier, can cause many problems. I liked milling composite plates in a water bath, and this isn't an option with my new cnc, as one of the axis is realised by moving the table. But so much performance will outweigh the struggle with the air filtration.

Hey Hanaldo,

I happen to own a Stepcraft 480 Black. It depends on what you want to do with your machine, but for my purposes it is good enough as an entry machine. I have a lot to do with 3D-printers (FDM, SLA, SLS), and nothing 3d-printed (even 50k$+ machines) rivals my Stepcraft considering part accuracy and reproducibility. And I am pretty sure if you only need your machine once in a while for some epoxy tooling board milling and you don't need sub 0.1mm accuracy, an X-carve will do just fine. Even if you plan on eventually investing $5k+ on a cnc machine, I would always recommend to play around with an X-carve or Stepcraft first for a couple of months to get the hang of everything. You first need to understand feeds & speeds, CAM, different bits for different purposes, the right spindle (i.e. brushless) and how everything comes together. It really depends on the size of your projects. For example, I would need maybe 8-10 hours of milling to produce the mould in the above video, simply because it is too big and I can't be more aggressive with my feeds and speeds, my spindle, and am also missing stiffness in the portal.
CNC Forums can be a nightmare if you are new to CNC, lots of contradictory information. That's the main reason why I would recommend a cheaper entry to see for yourself if it is for you and later — if necessary — upgrade to something with a little more bite to it.

@oekmont looks impressive. Where are you located? 

Very nice oekmont, certainly far more than I would ever need for the size of my business as it is! Very envious though, I'd love to play around with some serious equipment.

Cumberdale, thanks for that mate, I appreciate that advice. I did wonder what you were using for working with your tooling board and had hoped you would chime in, so I'll keep that in mind. I think the initial excitement has died down a bit now, so I'll be happy enough just playing around with the CAM simulations in F360 and seeing if I can get all of that right. Fortunately one of my customers actually owns a commercial CNC workshop, and he has all sorts of serious 3/4/5 axis CNC machines. He has offered to mill stuff for me if I do the drawings and things, so that may be a good way to start having a play and understanding how things work, I just always felt a little embarrassed to use a 5m x 5m machine to mill down a 150mm piece of tooling block! 😅

@cumberdale: a x-carve is a whole diffent story compared to a stepcraft. The working area is much bigger, while the profiles are much thinner, resulting in a much, much lower stiffness.

@ hanaldo: the feeds etc. Are dependant to the system you are using. For tooling board you are usually using a high spindle rpm, and then make the first passes as fast as your machine can handle. The finishing pass however is made much slower, with a ball end router bit. The smaller machines usually only take small bits, resulting in a smaller radius of the ball end, so you need a far more passes, to get a good surface.
Remember, that an x-carve is suitable (not ideal in my opinion) for cutting out 2d shapes. So plug slice cutting could be an option. This could make things a lot easier and could be a nice cnc intro. You should certainly not start with 3d anyways.

Yeh I considered it for that purpose, but I think if I'm going to spend $1500 on a machine (as well as dedicate a relatively large footprint of my quite small workshop) then it needs to do more of the job than what realistically only takes me 20 minutes to do by hand.

Really I wanted to skip the pattern making process altogether and go straight to machining small female tools for low volume production runs. Things like intake pipes for various vehicles, where they are so small and simple to do, but there's enough little variations between models that you would need dozens and dozens of moulds. If I could do that sort of thing quite quickly then it becomes more realistic to offer it for under 20 pieces, but otherwise I can't warrant making a mould unless I'm going to use it for dozens of pulls.