1. you can't just double the mold for the sake of math just like you can't just double a printplate. if you want more than 240x180 you'd need a second machine or double the time.
2. cast isn't print, the models are spaced out and horizontal so the space isn't directly comparable. in their sale video siocast had 32 models in a mold, but they didn't look heroic size so might be even less. with an area of 500x500mm and a generous 50mm² per model, one mimaki printrun nets you more in comparison.
3. casting still has a preproduction time, siocast says it can take up to 3 hours to create the mold, which comes after the time it takes to create the masters. after creating the mold up to another hour to vulcanize it. while waiting for all that the mimaki is already printing.
if I read it right at least the mold seems to hold 300-500 cycles, so that process would only to be done once for a while. siocast is gonna catch up eventually, but is an initially longer and generally more involved process. new/more molds could be done while the injection is running for the first one, but would still have to be done for a different model range. inkjet you can print whatever all the time.
4. the post processing machine is optional (I assume it's just a beefed up ultrasonic cleaner to industry standards), the mold press to create the mold is mandatory and would add to the siocast price (and space), plus a normal 3d printer to create the master and runners.
5. casting still puts a physical limit on the model for the process which additive manufacturing doesn't have. this mainly depends on your models tho.
6. not really relevant but inkjet printing in color is one process, painting the cast would add to the overall time.
price-wise for easier comparison the 2207, it has a builtplate of 203 x 203 x 76 mm, "roughly" the same size (it's funny they don't mention times for that model, I assume it's the same), but that one only costs around 40k. siocast according to the prices I've seen is 50k € for the sioform1 and another 10k for the press (doubt you really need an additional SLA/SLS printer since it was a formlabs article, a regular consumer 8k resin printer would be peanuts in comparison).
that's without prices for material, who knows what siocast or mimaki take for their stuff.
with that in mind it depends on the volume and your models over all, speed is secondary if you have the time and the printer sits idle half of it. not that you'd plan to if you're willing to spend that amount of cash....
or if you're lazy like me inkjet is simpler since it's just press print and scrape it off the plate after after a few hours. ¯\_(ツ)_/¯
1. Sure, and you'd actually need to square it, but that's not the point. Also, you wouldn't need to double the machine to double the area, when the siocast machine handles it in 60 seconds vs hours if you're talking purely about output. But I do mention simply getting 2-3 machines due to the machine cost and space they occupy.
2. You're right, however more models in one print is going to be more post production time. Also, even if they were heroic size 28-32mm models, 32 models in one print if it takes 2.5 hours to do, vs how many that could get done via siocast? Even if you only got 8 on a sprue, that's still only 4 minutes of machine time to produce 32 minis at that point. 1 operator could easily handle 2 machines, and 4 molds to swap back and forth. Even if we count the 4 molds as having taken 3 hours each, that's still 12 hours to have made the molds plus an additional couple of minutes, the moment your production gets beyond that initial half dozen uses per mold, or a few minutes, you'd already be ahead of the mimaki.
3. Yes, 3 hours to create the mold. But even if you only get 300 shots out of the mold(which btw, you'd want 2 to maximize time for one siocast machine anyway) you're still ahead. The mold is processed in a different machine, not the injection machine(which really means the cost difference is going to bring that closer to having 2 injection machines not 3 for the cost of the mimaki but... that's also not including the rest of the mimaki equipment such as the post processing machine that I mentioned either, so it's kind of moot).
4. Getting the mimaki specific post processing machine is likely optional, but that doesn't change that you need a post processing machine. Even with a generic machine you're adding a lot of time to deal with the parts, plus inspection, this is human time, which costs a lot of money if you're not operating out of China or some equivalent.
5. This entirely depends on the model.
6. Yes and no, the mimaki makes this one process, but you're still paying for it. Hell, it likely does cost as much as inkjet compared to most mass production painting techniques because you're still paying for materials.
And sure, comparing it to a different mimaki machine if you want to just look at build area alone could make sense in some regards. But what's the print time for the area? That is going to be a huge consideration. The fact is, 2 injection machines, with 4 molds is going to be much faster than a couple of mimaki machines, period.
You also need to consider other aspects. Packaging of flat sprues vs packaging of pre-assembled 3d models. How much does that cost, how much does it cost to ship? How much shelf space does the same amount of product occupy in a retail environment when it's packaged as an assembled model vs sprues in a flat box? At that point a comparison to flat pack furniture can be made, and Ikea has managed to make that into a massively profitable business model. Even warehouse space costs money.
Then getting into just the fully assembled and painted product vs hobbying... the fact is some people actually like the hobby aspect. Doing the work from the factory doesn't necessarily eliminate that. People obvious stripped and re-painted their own stuff for x-wing. But then you're also simultaneously limiting revenue from the auxiliary products as well. GW, Privateer Press even. Paints, brushes, bases, cement, knives, clippers, etc. are an additional revenue stream.
The simply reality is, if those mimaki machines actually worked out for production at scale for a miniature company, we'd already be seeing it. People get paid quite a bit more money to do actual analysis vs some incredibly quick and admittedly halfassed napkin math that we're doing here. Hell, those people are going to take into account labor cost, cost of PPE or other safety equipment required for handling and production, the energy cost to run the machines, possible leasing out of time on machines(companies do this all the time in a lot of industries) to more quickly amortize the cost or bring on additional revenue, and so on. I think the fact that we're not seeing them used when they've been available for a few years now speaks for itself.
Even for lower production run products, there aren't many companies doing this. I mentioned Weta Workshop, Sideshow, and Gentle Giant and others. But these are companies doing 1:6 scale statues, 1:2 scale busts, etc. at low production volumes. Only doing hundreds, maybe a couple thousand pieces each, and they're still charging hundreds of dollars for them. I'm assuming the production run quantity and pricing has been checked out by someone analyzing all of the costs to make sure those companies can turn a profit.
And sure, for you or I, scraping something off of a build plate is going to be quicker than any of the additional fucking about required for siocast, pouring resin casts, and so on. But the problem still arises when having to do it as a business at the scale GW does. They could probably get away with it for forgeworld stuff. Just do monocolor on a mimaki for things like the horus heresy primarch models(hell, even just porting over models of the existing sprues really) due to the low sales volume and high consumer cost of those minis. But for most of the product range? If it made sense, they'd probably already be doing it(especially since all of their modeling is done digitally now, and avoiding the tooling costs, mold switch out times, mold storage, etc for their current processes could be eliminated).
edit: And no, I've got nothing against 3d printing. Hell, just recently bought a new FDM printer.
