the robots are coming

[estimated reading time 19 minutes]

one of the most frequent questions in woodworking in the modern era is “should i get a cnc?”. the answer, as with everything in contemporary crafts, comes down to personal perspectives. if you’re a recreational woodworker not looking for serious production capacity, the answer is likely going to be no. there’s very little you can do with a cnc that will benefit you and the cost of entry is rather high. if you’re an individual maker looking to compete in the market with your products, though, it might be exactly the edge you’re looking for and if you’re in a cabinet or production shop with a few others, if you’re not using a cnc you’re probably missing something.

most of you, however, are on the hobbyist end of the spectrum. so what about you? is a cnc something you should be devoting serious investment, time and space in as part of your woodworking life or is this a technological development you can ignore as yet another mass-industrial movement that might do more to destroy handmade pieces in the public imagination but doesn’t really have an impact on how you work? let’s take a look at some of the basics. i’m not going to tell you if you should buy one. but i’ll do my best to make the decision process a bit easier for you — there’s no substitute for information. except perhaps misinformation. and we have the internet for that.

what’s a cnc?

cnc stands for “computer numerical control”, which is, other than grammatically-spurious, completely meaningless unless you already know what it is. these machines are computer-controlled devices that cut and shape parts. the easiest way to think of the type you’ll see in a woodworking shop is that it’s a handheld router on a robot arm so instead of moving it with your hands and jigs it’s moved by a computer. there are more-involved versions including lathes and rotation but for the kind you’re likely to see in a small shop in the early twentyfirst century just think of a robotic router — anything you can do with a router, you can do with a cnc. if you can’t do it with a router, likely not going to be possible (definitely not easy) with a cnc.

if you’ve never seen one in operation, think of it as something like a drill-press with a table that can slide side-to-side and front-to-back but instead of only drilling a hole with its spinning bit it’s got a router bit so, as the table moves and the head moves up and down, it can cut lines in the piece. this is a bit of a rudimentary description but it’s a good mental picture. if you’ve ever used a pattern shaper, that’s a good way to think about it except it works in all three dimensions and you don’t have to control it manually. what this leads to is significant improvements in precision. but they come at a huge cost.

how much does a cnc cost?

the simplest way to figure out if you’re going to start using a cnc is to determine if it’s even possible within your budget. there are a few entry-level cncs out there — and i’m going to tell you getting one is probably a bad idea except in a very small number of circumstances. [inventables makes a commonly-video-present one called the xcarve] with a base that’s a little less than a square meter and runs on a tiny dewalt router. it works fine and is the only entry-level cnc i can honestly recommend but it comes with a lot of caveats (more on that in a minute) and it’s about two grand (or equivalent in local currency). not to mention you pretty-much have to build the thing from parts when it shows up, which is a huge amount of setup even compared to what you have to do when a sawstop shows up at your workshop (yes, for that kind of money i expected to be able to turn it on, square the fence and start cutting and that’s absolutely not the case — a half-day to set up a saw is a bare minimum).

my other recommendation for cncs is to [go with avid and they have entry-level kits] starting about the same price (yes, i’d get the entry-level kit from avid rather than the xcarve given the choice) and much the same level of construction requirement. from their $1975 600mm x 600mm model, it quickly steps up as you get larger, more useful tools. the first one i consider particularly-useful in a workshop is their $2725 model with a 600mm x 1200mm bed but if you’re going to be doing significant work with it you probably want either their basic 1200mm x 2400mm machine ($4625) or their “pro” series in the same size ($6150). i won’t go into the details on these — though i certainly can talk about the pros and cons of avid and their specific machines if anyone is seriously considering getting a cnc. but that will give you a pretty good guideline on the cost of getting started. here’s a breakdown.

  • you’ll need the machine — $2000 to start, potentially $3000 to get something more useful or double that for a serious piece of equipment to save you a lot of time in the shop.
  • you’ll need a computer to run it — if you think you’re going to be running your cnc from your laptop, please think more carefully about this. for one, do you really want to get your laptop that close to all the dust in your shop for hours at a time? second, the thing has to be connected for the entire time the machine is being used and this is a cumbersome and annoying process. every time. so, while this isn’t a massive expense, assume you’re going to be getting a dedicated computer to run it. it can be an old laptop you pick up for a couple of hundred bucks or an old desktop (does anyone have one of these anymore?) but the key is you probably need a place to put it and this expands the footprint of the machine.
  • you’ll need software — if you’re a hobbyist, you can use fusion 360 for free and that’s exactly the software you want but if you’re using it in a production environment (as in, if you’re selling the stuff you’re building with it) you’ll need a license for fusion 360 and that’s about $500/y (it varies by location but not much).

with those three things, you’re ready to get started but that’s not a trivial bit of investment capital. if you think you might have the money, though, the investment really does have the potential to pay for itself.

what’s a cnc for?

there are three main uses for cncs (i’m using this in the woodworking sense — i’ve used ones meant for metalworking and industrial uses but this isn’t relevant so just assume i mean things like the avid and xcarve, not the ones made by mitsubishi and baker — these aren’t useful in your shop, as much as i recommend looking at some pictures just to get an understanding of what the other side of the mass-production garden is looking like — it’s not nearly as green as you might think but it’s definitely shiny and anodized). flattening, templating and automation or batching. let’s take a look at each of these individually.


one of the things people rarely think about when they start contemplating a robot shop-helper is milling boards but that’s perhaps the most useful thing a cnc can do for you in a traditional woodworking shop. how much of your time is spent getting a board flat and square? a lot? this is work you can outsource to the robot and it’ll do it better than the place you buy your lumber, i guarantee. the limiting factor, of course, is how big your cnc is. but if you rough-cut your stock to length before starting the milling process and generally don’t make pieces with huge components, this might be a very smart idea.

here’s why it’s economically-viable. you don’t need a jointer. so you can take your entire jointer budget and spend it on your cnc of choice. why doesn’t that matter? well, if you read my article from a few weeks ago about milling without a jointer, you might not have a jointer budget at all because it’s probably the least-useful tool in the shop unless what you’re aiming for is a huge amount of production. if i was going to get a jointer, though, i’d skip the jointer and get a 1200×2400 cnc to do the work for me. i hate jointing (by either machine or hand) and this simplifies things immensely.

what’s the real upside to this? slabs. if you make slab furniture with huge pieces of wood — perhaps a 1m x 2.5m slab, for example, as a tabletop — you probably don’t have a planer and you’ve likely been flattening the thing with a router-sled and a flattening bit. this is messy and slow. it certainly works but it’s a nightmare if you have to do it repeatedly. make one slab table and it’s ok. make three a week and you’re probably starting to hit your limit in terms of router-jig patience. i don’t blame you. the only thing to keep in mind about this is that you’re going to need a cnc big enough to hold your workpiece. you can certainly cut it to rough length before you mount it on the bed and turn it on but if you want a 2.4m final length on your tabletop you might want to seriously consider a 3050mm bed for your cnc (yes, this is available as an option on that avid pro-series model i mentioned earlier).

so if you’re a frequent producer of slab furniture, a cnc might be a no-brainer for you. it will save you hours on every project and mean you can probably double your output. will that pay for the cnc in a couple of weeks or will it take a few months? i’ll let you run the numbers. but it’s likely worth the investment on the flattening front. and once you have it for flattening you can use it for all the other things it’s good for.

if you don’t make slab furniture, you can certainly use it to flatten regular lumber — it’ll even work for green things if you’re patient and careful with your settings. if you’re using standard dimensional-ish stuff from a lumberyard or hardwood-dealer, though, you can probably fit that stuff in your planer and the advantage in either time or annoyance will probably be far more minimal than in slab-land.

some of the questions about flattening, though, really do come down to money. could you sell a bunch of slab furniture if you had a way to make it quickly? is this somewhere you want your woodworking business to go? is anyone doing this in your area or would you be tapping a niche market with hundreds or thousands of potential customers? slab furniture is relatively simple to build with minimal complex joinery — most of what people love it for is the look of the slab and that’s more a question of sanding and finishing what nature gave you rather than having to build anything. so if you can knock out a few slab tables a week could you bankroll your woodworking venture? would it make the purchase of a large cnc worth it? maybe. these are questions everyone has to answer for themself but it’s definitely possible — some will depend on your area, market, desires, etc. do you like producing this kind of furniture? would it be worth doing it to make other things you enjoy more? also important questions and i’ll leave you to ponder them.


this is what i think the most practically-useful task for a cnc in a small woodworking shop is. not making parts. making templates. if you’re not using templates already as part of your daily workflow, this might be something you should consider. many of the things i enjoy making are made many times. and they have complex curves and angles, angled joinery and some serious compound angles. cutting those things once is hard. cutting them the same every time is almost impossible. but shaping them with a variety of tools from drawknives and spokeshaves to sanders and scrapers from pieces of plywood or, often, hardwood is actually surprisingly easy. especially if you’re making templates for segments of components rather than entire components. then you can attach the template to the roughly-shaped piece of wood and use a router with a template (also called a flush-trim) bit to get it exactly right. this is a huge part of my in-shop procedure and i highly recommend it if it’s not something you’ve tried.

there’s another great advantage to working that way, too. i do all my design in fusion 360 — yes, sketchup is unintuitive and i suggest you avoid it, especially if you’ve never used it — fusion is easy to learn and you should watch lars christiansen’s absolutely amazing livestream videos to get a good handle on it before you even open the program the first time. enough doing autodesk’s marketing for them, though. just get fusion 360 — as a hobbyist, it’s free. that means i can generate plans and print them lifesize in pieces then tape those together, stick them to the wood and i have a paper template to make a wood template from. once i have a single wood template, i can make an unlimited number of copies from the final stock. not to mention it makes prototyping very easy. print on paper, cut from 3mm cardboard or door-skinning material (yes, this is extremely cheap and you can cut it with just about anything — butterknife? teeth? feather? no problem) and stick it in the proposed living-space for the piece and see if it works. wait, you’re not making mockups before you build your pieces? barbarians! i’d have made so many inappropriate pieces of furniture that didn’t fit their spaces or suit their target users if i didn’t mockup. sometimes i do a few before i settle on a final piece. if it’s not mocked-up, though, i don’t let myself touch the real wood, with very few (and extremely simple) exceptions.

what does that have to do with cncs? if you’ve already made the shift to template-driven woodworking, instead of it taking hours with a printer, scissors, tape, glue and a dozen handtools and sanding machines to get your templates right, it’s a few minutes to go from fusion model to actual, usable template. you can think of this as the equivalent of printing a document — you literally print your templates on wood and, while a little slower than a printer on paper, you can often have the piece you designed in a matter of minutes and it will only require a tiny amount of sanding before you can use it to produce dozens or hundreds of functional parts in real wood.

there’s one other component to this, by the way. do you design furniture and sell your plans? how much more valuable would they be if you could sell actual templates with them? if that’s something you want to be able to do, there’s no substitute for a cnc. just stick the template material (my arch-nemesis mdf, usually) on the bed, tell it to cut the templates and walk away. you can batch them out or you can just have a single set of templates for a particular project and each time someone buys one literally just print a set and stick it in a box to take to the post-office. is this a use-case for everyone? not even close. do some of you already make plans and sell them and wonder if template-production is a way to up your game and profits? certainly enough of you to be worth mentioning it.

if you haven’t tried template-driven woodworking, though, especially if you’re like me and work particularly in styles related to the danish-modern movement with flowing curves and complex angles, this could be a game-changer for you regardless of whether you actually buy a cnc. if you’re wondering whether this could be a good reason to get a cnc, try actually making the templates by hand and using them. if you find it useful and want to save all the hours of templating, a cnc might be for you. if you don’t enjoy this procedure, you’re not going to like it any more when the templates are being produced more quickly. some people do this craft for the end result, the design and the furniture (this is why i do it) and others do it for the procedure. if you’re like me, this is awesome without question. if you do it for the procedure, try the procedure and see if it brings you joy. if it does, keep it. if it doesn’t, maybe what you need isn’t a cnc but another drawknife and a new shavehorse. and that’s so much fun, after saying it, i’m probably going to go and drool over that [boggs shavehorse] again and wonder when i can get into my new shop so i can build one for myself. (yes, i designed a shavehorse and the plans are floating around free somewhere but it’s a basic one and the boggs one is awesome af if you’re not a beginner.


what most people think of when they think of getting a cnc is actually the least-useful task if you’re running a small shop — it’s the most-useful for a large shop producing hundreds of cabinet parts or furniture components but that’s not most of you/us. producing parts reliably and repeatably is what robots and computers do best. and usually what humans do worse — we don’t have either the attention-spans or physical prowess to do it and handmade things usually look imperfect. this can be a blessing or a curse. some things that aren’t perfect are absolutely beautiful specifically because of this imperfection. other things (come on, people, sand the entire table, not just the visible surfaces — every time i see a table with the bottom left rough because someone thinks it’s aesthetically-pleasing to give me splinters when i hit my knee on the bottom i just want to burn the thing) not so much. if you’re only making ten tables a year, they’re probably all going to be different. batch-processing isn’t really all that useful. if you’re producing ten a week, it’s suddenly far more useful. producing ten a day and you’ve already got a cnc or an army of helpers and you’re not likely reading this article anyway.

there are a few places automation and batch-processing are useful for even the small shop, though. are these worth getting an expensive and massive (yes, i can’t emphasize how large these things are and we’re talking footprints multiple times the size of your cabinet-saw or floor-model jointer) machine? i’ll leave that up to you. chair parts are precise and often repetitive. case components, especially for things like bookcases, can often be batched. make a lot of cutting boards? bash those things together and stick the rough stock on the bed and you’ll get them off ready for finish-sanding and their bath in shellac faster than you dreamed possible. batch-production can be extremely useful. if you make a lot of the same thing or pieces with many of the same or similar components, a cnc could save you hours or even weeks of effort. if everything you make is highly-customized and there’s rarely duplication of shapes or components, this won’t do you any good and having a robot helper is a pipe-dream with a particularly-hallucinogenic pipe.

again, though, this might be worth saying. are you having difficulty making enough profit to support the kind of furniture you really want to be making? is there a market demand in your area for things like engraved cutting-boards or simple chairs/tables/bookcases you can batch out on a cnc? this might not be the investment you want to save for to make your woodworking career better. this might be the solution to your cashflow problems. maybe. but it’s not a panacea and you have to do your research. you’ll have to sell a shitload of $100 chairs or tables to pay for a $4000 machine (about a hundred if your profit margin is the usual 40% for such items) — do you think you can sell ten? a hundred? a thousand? do the math. do the research. buying a new carcass saw without market research is totally awesome. just get one. [this one, in fact]. you’ll love the living fuck out of it. buying a tool the size of a sedan to park in your workshop that’s as expensive as, indeed, a used car? know what you’re getting yourself into before you pull the trigger on that. or you’ll be seriously thinking about other uses for that shotgun in the corner.

what about handheld cncs? budget cncs?

i know what the next question is going to be. you’ve probably been thinking it all along. you don’t have the space or the budget for a 1200×2400 avid. you’re not even sure you want a 600×1200 to make templates and flatten smaller stock. but a company called shaper makes a handheld cnc router called the origin and you’ve seen it on youtube dramatically-endorsed by your favorite influencers. and some of these are awesome woodworkers. and you’re not sure but you think you might want to buy one. don’t.

remember what i said about the [jointmaker pro when i reviewed it and got a fair amount of flack for]? that it’s the most useless and overhyped tool in the history of small-shop woodworking? well, i stand by that. but second on the list of overhype is the shaper origin. in its defense, it’s a well-made tool from a company that is market-aware. on the other side, it’s ridiculous. and this is why — it removes both whole points of getting a cnc — it adds consumables and a user.

let’s think about why you might want a cnc. use it to flatten a slab while you do other things? perfect. use it to cut out templates while you prepare the stock to apply the templates to? again, what it’s designed to do. batch-cut parts while you go looking for clients to sell them to? three for three in favor of the cnc. except the origin can’t do any of these things. because not only do you have to use its (admittedly-ingenious) tape to show it where to cut, it’s incredibly slow (by necessity because it’s stabilized but humans are shaky af) and you have to actually be there every step of the way. it’s a robot-helper without the … you know … robot. it can’t move itself. you’re the robot part and all it does is cut. so realistically this overhyped waste of money is a handheld router with its own self-correcting jig. and while that’s an amazing technical and technological accomplishment, i can’t see any reason it would be useful in practice.

if you’re a tech nerd and you like gadgets, this is your next one. it’s brilliant and perhaps one of the most impressive pieces of tech kit i’ve seen. but will it actually give you the advantages i talked about in the three main uses of a cnc in your shop? absolutely not. it’s a toy. you get into woodworking for the toys? get a shaper origin. you get into it to build furniture? perhaps even sell furniture? go big or go back to the basement-shop.

there are certainly companies selling budget-cost cncs, too. often ones that are combination laser-cutters and routers. while their lasers are sometimes usable (though rarely more than barely-usable), their routers are pathetic. i have yet to see anything cheaper than the entry-level avid (about two-grand) that was worth even contemplating. and there are hundreds of them. you can do the research if you like. but i suspect you’ll spend many hours finding the same thing i have — they’re, loosely-put, a group of objects as useful to woodworking as a mound of excrement on your shop floor. not only do they not do what you want, they take up usable space and fill you with disgust every time you go near them. you can get by with a budget chisel or a twenty-buck prewar stanley. buying a budget cnc is like getting one of those molded-plastic chairs and using it at your dining-table. will it get the job done? maybe. will you hate yourself for it? immediately.

what’s the real deciding factor?

that’s a bit of a trick question. there are two — do you like the cnc way of working and is it a good investment for you. they’re totally different questions but let’s summarize them.

feel-good wood

do you like designing? designing on a computer? cnc might be for you. do you like working with templates and complex curves and angles or are you more into rectilinear forms? if you’re down with the squares and boxes, are you doing that because you’re limited by skill or is that the aesthetic you prefer to work with? would you like a helper in the shop (robotic or otherwise) or would you prefer to do everything by hand? there are relatively few handtool-only woodworkers out there. most of us are on the spectrum of more-or-less-hybrid (i love handcut joinery but almost everything else is done with electrons in my case) — you probably use powertools for some things and handtools for others. but if you don’t like working with cad software, a cnc is not for you. just don’t even contemplate going there. you will spend hours in front of a computer and if that fills you with a limitless foreboding of dread and aggravation this is a world you don’t want to inhabit. if you think you’ll just hit “flatten” and get beautiful slabs ready to handcut mortises in, think again. it takes serious time and at the beginning this is serious time investment with little to show for it. the investment pays off in being able to do things much more quickly in the future but that only works if you actually want to do this type of woodworking.

the simple question is do you like jigs, electronic or otherwise, robotic, human or seat-of-the-pants-from-plywood? if you’re not into jigs, the cnc is the mother-of-all-jigs and thinking of it any other way is probably a mistake. if you’re into pushing the boundaries of what you can do using whatever modernized tool you can get your hands on, cnc might be a good fit — but (and this is important) this is a world you have to dive into. you can’t dangle your feet in the water. get a serious cnc, not something from ebay dropshipped from a warehouse in shenzhen for five-hundred-bucks with a laser attached. yes, there are awesome tools made in china. and there are cheap tools made in china. there is no overlap. and there are plenty of shitty cncs made everywhere including right at home in america. spend the money, spend the hundreds of hours to get up to speed in fusion (or something else but, seriously, there is nothing else worth the effort and you should be using fusion — those people whining about easel and aspire and such are really just … well, they don’t know what they’re talking about if you want to get serious about this and if you’re making this kind of investment you want to get serious about it, don’t you? using fusion is like buying a good quality saw and learning how to use it — the free software to run the cnc is like trying to jerry-rig a dewalt contractor’s saw to cut precise dovetails then thinking you might be able to use it as a chisel if you can just figure out how to hold it right) and you’ll be rewarded. if that’s where you want to do your woodworking. if it’s not where you want to be in the first place, walk away and don’t look back.

robot woodworking either makes you smile or it doesn’t. if it doesn’t, that’s not a bad thing. just don’t waste your time. if it fills you with joy to think about it, though, this might be a place you want to dive into. might be time to get in head-first. but don’t do it blind.

financial fuckups and freedoms

cnc is expensive. you’re going to spend at least a couple of grand on the equipment and if i was starting out now i wouldn’t even consider anything that couldn’t cut at least a 2m piece in one direction because so many components i want to template are bigger than a 600mm or even 1200mm cutting window on one axis. yes, you can get started in the 1200×2400 game with an avid complete with a full license of fusion for five grand. and that’s about the same cost as a 5p sawstop industrial-cabinet-saw. so it’s not outside the realm of possible but it’s not trivial loose change or even bad-axe-carcass-saw territory. this is like buying a used van to deliver your work — an investment in the future of your woodworking business. and you should probably think about it that seriously.

on the other side, having a cnc might mean you can do things far more efficiently — slab furniture, template-batch processing for complex pieces, automated production of simple objects. the other thing you might be able to do is rent time on your new machine to other local makers and it could pay for itself in no time — or it might not, if you don’t have any demand for it. cabinet and sign shops do this all the time and some invest far more money than is wise hoping to recover it from rentals that don’t appear while others get one machine and find there’s so much demand they never actually get to use it at all and have to buy three more just to keep up with the small rental clients who don’t want to get their own but come in once a week to borrow the thing for a couple of hours.

is it a good financial choice for you? nobody else can tell you that. should you do the research and be smart about it? absolutely. but finance alone won’t tell you if you should buy one. you can probably make a lot of money if you get a van and offer careful moving services for furniture, too. but is that what you want to do for money? maybe. you can produce lots of furniture more quickly with a cnc. is that worth the investment? again, maybe.

final thoughts

cncs are wonderful. i’ve loved them both conceptually and in practice for decades — actually, since i was a teen in the nineties and first got into cad modeling on a primitive machine with autocad i’ve been completely hooked. is it right for you? imagine spending three hours this afternoon designing a piece of furniture on the computer then three more hours this evening cutting those parts out on the cnc and assembling it. if that intrigues and stimulates you, welcome to my world. if that sounds like as much fun as digging your own grave then taking a nap in it, you also have your answer.

but the biggest bit of advice i can give you after years of playing with and thinking about these things is simply this — get serious about it or walk away. go big or go home. cnc is an amazing world to be in but you can’t sort-of enter it at the bottom end and hope to play either cheaply or without time investment. it’s a huge shift in the way you do your woodworking and probably in how you schedule your time. now it’s your turn. what else do you want to know about digital fabrication in wood? templating? slab furniture by robot? i’m all ears. thanks for reading!

share on social media...
thank you for reading. your eyes have done me a great honor today.