Introduction to CNC

This conversation, triggered by a simple open question, amounts to a nice overview of the decisions and experiences involved in transitioning a shop to CNC production. October 20, 2013

(WOODWEB Member) :
I read about so many software programs that you are all using for cabinetmaking businesses. Do all CNC nesting routers talk the same language, or do I have to shell out piles of money for software to run a Biesse, Thermwood, Onsrud, etc.? Can a guy write his own G-code, or is that stupid? I hear tell of companies that have paid 30k for software to run their CNC.

Forum Responses
(CNC Forum)
From contributor D:
Yes, software that runs CNC can be expensive. Yes, you could write your own gcode to run the machine but it would be very time consuming. I run Cabinet Vision and paid them about $20k for full screen to machine for 5 seats, onsite integration, and the post to be able to output to the machine. I spent an additional $10k over the years to a specialist to set us up and get everything perfect. I can now take a project and output to the machine in minutes. I know what gcode is, but couldn't write a line of code if I had to.

From contributor J:
You may not find a "one size fits all" answer here. I would step into it at whatever path is comfortable without breaking the bank. CNC can be the greatest step one takes, or the worst decision ever made. Go see some shops with machines, listen and take notes. I met with 2 CNC machine companies. Go to mid size and large companies, call CAMaster (the one I have), call Shopbot, Thermwood and all of the ones on your list - they will tell you where owners of machines are near you. Do everything you can not to buy anything until you are absolutely comfortable. I started looking and purchased 6 months later.

From contributor B:
Just about every machine has a different g-code formatting, so writing code to different machines from the same file requires post processors to format the code correctly. You can't simply take code from one manufacturer's machine and run it on another without some editing, in most cases.

The type of software you'll need depends on what you're trying to accomplish. If you're creating a lot of complicated or unique parts, you wouldn't want to have to program all of that manually. But if you were only making some simple drilling patterns, etc., manually programming is entirely possible, and in some cases preferable (parametric patterns, etc.).

Many machines can read dxf files and create code from that. There are many budget (and even free) software options out there, but if you want everything working out of the box with all the bells and whistles, you may have to spend accordingly.

For making cabinets you should check out Thermwood ecabinet systems. This software only works with Thermwood and Shop-bot machines.

From contributor F:
You may want to check out cabinet parts pro.

From the original questioner:
So let's say I have Cabinet Vision. Will it output to any CNC, or only to specific? I use ecabinets now but I know it interfaces only with Thermwood. I was thinking of trying to get a used machine for half the cost of a new one, but then will run into the software problem. Maybe used is not the way to go, as there may be many issues with an older machine. I would not mind a used Thermwood, though, as I am comfortable with the software. Just saw an SCMI Pratix go for 20K.

From contributor D:
Yes, Cabinet Vision can output to multiple machines. It does require a post processor unique to each model of machine. If you have a Biesse Rover and an SCM, it will output to both but you will have to purchase a second post and choose which machine you will be sending files to in order to create the proper code.

From contributor B:
Cabinet Vision will output to many machines. Not sure if you have to pay for the machine links, though. 20k for a Pratix is a good deal, assuming there is nothing major wrong with it like the spindle is shot or something.

From contributor F:
You can output your ecabinets with a link to a Shopbot.

From contributor J:
Mozaik CNC is a real option for anyone wanting to make cabinets, whether with CNC or not. No radius walls yet, but they have not ruled it out. So far I am getting into it.

From contributor C:
If you buy a used machine, look at the software on board with the controller. Some of it is pretty powerful. I know someone at a software company is taking classes at Biesse to learn their Biesseworks software, because it is so incredibly powerful. My controller has the software that it is based on, and if we didn't need shop drawings, I never would have bought Cabinetvision and the Screen to Machine module. But being the bonehead I am, I thought we should just field measure, verify and send to the router. Yeah, very pricey, but paid for pretty damn fast.

So, you can look at the imbedded software, or Vectric is supposed to be very powerful. It should help you create some parts. Also the Mozaik looks great. Just dive in and stay the course to reap the benefits.

An SCM Pratix for 20k? Get on with it. Just don't stop production and count on it until you are seriously comfortable with it.

From contributor M:
If you are going to purchase a CNC flat table router as a nested solution, additional or onboard software should be budgeted as a requirement.

If you are going to purchase a CNC pod and rail machine to compliment, enhance or as an upgrade within your current process, depending on the control, you may not need to purchase a cam package. This is especially true if the onboard software has parametric capability (one program that can process thousands of different design panels by changing 2~3 variables within the program - which takes seconds) and you are in total control of product design.

For someone starting out with CNC, that approach can be highly profitable without the added expense or learning curve of software upfront. I have numerous customers, still in business in today's economy, by using the right mix of CNC technology for their business model! One size rarely fits all!

From the original questioner:
Contributor J, if that is the menu that you see in Mozaik software, it is identical to Sketchup. They must have some parametric parameters to it.

This CNC stuff will be a lot of research before I purchase. I want to go to a shop where they use one and see how it is done from start to finish. I recently got a quote for a Thermwood 43 and whew, it would take a lot of kitchens to pay for that!

From contributor J:
Check out my new machine, under 45k with vac and atc, made in Cartersville GA, and has a heck of a support forum,

Click here for higher quality, full size image

From contributor C:
Had the same reaction on a quote from them. Bought a used 5x12 Rover 30, screen to machine module from Planit Solutions, a dust collector (on e-bay), pipe, a new air dryer, a used edgebander for backup, a tooling package and have money left over.

From contributor H:
I just bought a used 2000 Weeke BP60 from a State of Kentucky high school program. 118 minutes on the clock. Changed the cmos batteries in the computer and controller and the Stiles tech suggested I replace the vanes in the pump.

Also replaced all the pneumatic lines - 1 hour and 20.00 worth of 4mm hose. Had Stiles tech out for calibrating and training - 8 1/2 hours.

To my surprise it fired right up and runs perfectly. The Weeke software is easy to learn and, from my very limited knowledge, will:
Net parts to size with router, do all borings, and saw a groove for the back, in standard and mirror image.
Do curved routings for arches and toekicks.
Weeke software stores the programs on board, making recalling parts for machining quick and easy.

My budget is very limited and I have been researching and looking for about 6 months. It is a pod and rail machine but I do confirmat or dowel, so nesting wasn't really a requirement for me. Never thought I'd be running a CNC, but for 3268.00 I am very pleased with it. For what it's worth the machine was missing two vacuum pods and has a couple of light paint scuffs.

From contributor R:
Lots of good advice. Write your own gcode? I run some 3d files that are over 200,000 lines of code. Surfcam spits that out in about 15 seconds or less. Got a hunch that pays for the expensive software real quickly. I'd shoot myself if I had to hand run that many lines of code.

From the original questioner:
There is the thing - if I see a used machine come up on an auction, how do I know what software drives it? I guess I have to find, say, 5 CNC manufacturers and know what it takes to run the machine. I don't understand how the hardware and software choices go together. In the end all I really want to do is nesting and cutting cab parts for now, but I don't want to overburden myself with a huge purchase in case the economy tanks here, which it will sometime.

From contributor D:
Most machines from Biesse, Holzher, SCM and others made in the last 10 years come with an on board software. The ones like this that I am familiar with will allow you to program a single part at a time. If it is programmed parametrically, you can change the size of the part and the operations will change accordingly. When you modify the depth of a cabinet side, you will also have to change the depth of the top and deck as the assemblies are not related automatically. We used our Biesse this way for about 2 years. If you don't change things too much, this will work well on a pod and rail machine.

The next step we took was to design with Cabinet Vision. This allowed us to set up constriction methods and parameters on how we wanted things built. When we changed the size of a cabinet, all of the code for the operations was automatically generated. We were able to create cultists for the beam saw and generate bar codes for the parts to be run on the CNC. The operator just scanned a barcode label and placed the part on the proper side of the machine. This saved a lot of hours of programming on a regular basis, although it did require an additional investment in software and if you are not great with computers, you will want to pay to have someone set it up properly.

The next step we took about a year ago was to go to nested base. I got rid of the pod and rail and put a flat table on our machine. This required a minor upgrade to Cabinet Vision to nest the parts and some more tweaking to get things dialed in. Now me or one of my designers can put a job together. My production manager checks it and we send it to the machine, the shop guy opens a file and pushes the green button and loads a sheet of material, steps on the pedal and presses the white button and stands back. In 5-8 minutes he unloads the parts, loads another sheet, steps on the pedal pushes the white button and repeats the process. In this process nobody has looked at code, programmed anything other than determining the size of the cabinet. It is much faster and more flexible than any of the above methods I previously used. Additionally we have almost no secondary machining operations on our parts. A week after we were up and running, I put my beamsaw on the market, and it was gone 3 weeks later, freeing up about 1000 square feet of shop floor.

A friend just bought a single user, nested base version of CV with a week of onsite integration and spent in the neighborhood of $10k. I went with CV because they were the only ones that I spoke with at the time that I felt could do what I wanted to. I am not familiar with any other ones at this time but others can do the same thing. Like anything, the more you spend, the more features you get.

I'll tell you this, though - knowing what I know now, I wouldn't even think of doing what I do now the way I used to do it, even if you gave me the equipment for free.

From contributor R:
Here's my take on software. 1) You need a CAD software to draw lines and arcs. 2) You need software to write GCode for those lines and arcs and then make a file with the beginning and ending language that the machine controller understands. This language gets the right tool in the spindle, sets spindle speed, tells it where to start cutting, how deep to set the bit, and then take it back to 0 and shut off the spindle. 3) You also have the controller software that stores programs, stores canned cycles like tool changes, zeros the machine, stores tool offsets, etc.

Some machines use proprietary software, some open source. Controller software comes with the controller, if it is a purchased controller.

Remember, you are just controlling a robot. The servos, or steppers on the CNC, need information to turn them on and off, and then how many revolutions to make. Seems simple, but of course it's not. That's why there are CNC techs, and if you can find an independent in your area, so much the better. Buying used with no support is not for the faint of heart. It seems to me that you are the kind of guy who needs a good dealer or tech for this project. A cheap auction CNC is no deal if it just sits there collecting dust. More spent up front will get you into production and making money, fast! I've been there.

From contributor H:
If I buy used or at auction, I look at the age of the machine, its overall condition, and what type products it was producing, but mainly its environment.

If everything else in the place looks abused and held together with duct tape/zip ties and in disrepair, that's a pretty good indicator of the machine in question. On the other hand, an organized company or institution that has the budget for proper installation and maintenance is a good bet for machinery.

If you can run it under power, fine, if you can't, then offer parts prices for it. If it works, when you get it installed, you're ahead. If it needs repairs, you have room to do so.

From contributor K:
You really need to take another look at the Thermwood CS-43. It is a fine machine and a workhorse. The shop I work in purchased it primarily for panel processing, which is its forte, but we use it for much more. I am currently designing some bow front vanities with integrated carvings that will be modeled using the 43 as well as a line of vent hood cabinets and refrigerator pediments. The best part is that this is all accomplished using the free eCabinet Systems Software that came with the router. I don't need a separate Cad/Cam package to design these pieces. I create the profiles in eCabinets then the Profile Modeler, which is integrated into the Thermwood control software, will model these parts using simple end mill and ball nose tooling. No specialty tooling required. The carving files are also integrated into eCabinets and the Thermwood control software. Using eCabinets, I create the part, model the part and add a carving. Nest the part and I am done. There are hundreds of carvings to choose from and they are all included free with your CS-43 and eCabinet Systems Software.

Some other features that come free with your Thermwood are their new Lock dado joint and the ability to create dovetail drawers. No extra purchases required. The "Lock Dado" joinery allows you to use the convenience of KD/RTA fittings with the alignment and quality of a "tabbed" blind dado.

Which software you use to run your router is probably more important than the router you choose. eCabinet Systems software is one of the most powerful design and manufacturing packages available today. It is capable of designing most anything you will need to manufacture your cabinets using any type of joinery, and because it is proprietary to Thermwood machines, there are no post processors required. A staff of Thermwood programmers work "in house" to constantly improve and update this software. You and your Thermwood machine are their priority 365 days a year.

There are cheaper machines, but there are not better machines.

From contributor C:
Cheaper, but not better? Please. Contributor J is running circles around high dollar machines that are 2-3 times his expense. I agree, Thermwood is definitely delivering the help with its file sharing, but don't discount the operators. I think Thermwood is a fine machine, but so is Camaster, and all the other machines listed.

Just go get a used or small investment machine and do what is comfortable. Contributor H's purchase is a prime example of low investment for a hellacious return and very low exposure.

From contributor K:
My point is that the cost of the machine must also include the software (and associated learning curves) to run it. In many cases this alone can turn a $45,000 bargain into a $75,000 one, depending on what you want to do. The Thermwood is a complete package. No hidden costs.

Contributor J does some great work but it is more his software than his machine. As I remember he uses several different packages to get the job done and you must add the cost of all this software (in money and time) to compare cost of the machines.

Thermwoods are great machines. You have one support call to make for the machine, the control software and the cabinet design/manufacturing software.

From contributor D:
I agree with you 100% - the software and how you are able to use it is very important. Software can't be looked at as an expense - it is an investment that pays a return. The whole reason for going to CNC is to do it better, faster and easier. Done properly, this move can save hundreds of hours a year in production, reduce the number of employees and their associated expense, and make you more money. If this can be accomplished for free, that's great, but even if it cost $20-30k, in a shop doing any kind of volume this would result in a savings of much greater than the initial investment. I would say the software is actually more important than the machine.

The used versus new debate is a whole other thread though. I don't know the original poster and his abilities, but I am going to make the assumption that his computer skills and understanding are not strong. He could save thousands on the initial purchase buying used, then spend thousands on a tech to come out and set up and make repairs, etc., have no warranty on the machine, and possibly no factory support on it, and have to integrate a software setup. Buying new normally includes a factory rep doing the install and software integration and being onsite for a week, leaving you up and running at the end. This has a value that can't be ignored.

Last thing, if you buy software, make sure you get it in writing what capabilities it will have, and don't pay for it all up front. This has saved me in the past. The software could do what the salesman said, but the tech didn't know how to implement. When you tell them "no works, no money" they are much more motivated when you are holding the money.

If I was starting from scratch, I would look at ecabinets and their wood for sure.

From contributor J:
Vcarvepro 600.00 (optional Aspire, 1995.00), cabinetpartspro 250.00. Other than drawing a plan on a napkin, this is the bottom line, and you could do anything I do. Yes, I have cabinetware and I design with it (actually most of our 50-150 k jobs are outsourced (1-1.5% of job cost). I draft basically all casework under 50 k with cabnetware (minutes) and enter box sizes in CPP for nesting. For cutting custom radius walls and parts I get the dxfs from my plans or the architects and cut directly from them with Aspire. I also have Cabinetvision (yet to really need it), cabinetpro and KCD (inherited them in a CNC computer and as I am not an owner, I do not use them (that would be unethical).

The video below is how one can use a 600.00 program to get started and still have plenty of use after one decides which cabinetry software path they will take. My goal was to make informed decisions with a minimum of risk. I love the fact that nothing I use is proprietary and I can work on and change parts on my machines without downtime (slave to no one). In addition all parts and support are minutes to hours away and support is free forever (no additional cost) and parts and factory fixings are free for two years! There is one other real cute feature of my machine. I purchased my x3 in March of 08; I sold it in October of 12 for 80% of its original value. Anyone seen a used CAMaster for sale lately? The time of turnaround used is usually less than a week. These are damn good machines.

On to Thermwood. There is not a bone in my body that does not recognize it is a well-respected CNC machine and in good company with many others. I have many friends who love them and work well with them. ecabinets though is also well respected by those who have devoted themselves to it. Contributor K is a well-respected designer with it. His work is first class using this software. In the two attempts I made to learn ecabs I found it difficult (it may be me). Having used CW since 97, I thought ecabs would be a cake walk (for me it was not).

Over the last few days with some spotty amounts of time, I was able to learn parameters (casework), design, optimize, nest and produce code with mozaik! At 125.00 per month (use it when you want and no contract), it costs as much as one foot of wall cabinet per month. It is powerful. I love the forum and the videos are online. I made a video last night and am having more trouble with the vid software than I am Mozaik.

My machine cost half as much as many here. The others have many features that after 5 CNC machines (including some bigger iron) I have found I did not need. On my new machine I opted not to get another lathe. Looks fun to play with, but for 90 + % of us, unnecessary.

My CAMaster ATC 508, WINCNC, pop up pins VAC table, etc. may not be the biggest, it may not be the best, but it does not stand second best to any other.

My biggest point should be, I am 100% against the pay us up front 20-30 k for software. This is old and antiquated thinking in today's economy. You can ease yourself into CNC with a minimum of exposure and potential loss, then make informed decisions as you grow into it. None of us have this capital to risk anymore.

From the original questioner:
This is great and I am understanding a lot more.

I was trained as a cabinetmaker and I then went on to become an architect and I am fully knowledgeable with 8 or so CAD and modeling programs, and was around when Fortran was the word of the day on the old IBM 4341 mainframe in the 80s. Seen lots of hardware and software since then so that area is not a problem. I just wasn't sure how the machine and the programs went together, as I have never seen them in use except on youtube, but now I am getting a better picture.

I know this is the way I am going to go to expand the business, I just don't want to hang out a lot of money and not get a good return in a reasonable amount of time struggling with a boat anchor.

From contributor D:
Contributor J, thanks for sharing. Mosaic looks very interesting and the pricing is great. I currently use 3 seats in CV but have 5 keys. I pay about $1500 a year in support, plus my initial investment.

The other advice I would give is to be open minded and don't try to force your current method of construction into the CNC environment, but look at how you can adapt the way you do things to the CNC environment. I had a guy that wanted me to cut parts, but only wanted it cut, banded and construction bored, but no line boring or hinge mounting holes on the cabinet or the door. He wanted to continue to do that on his machines and didn't or wouldn't understand that it would take the same amount of time to do it either way.

From contributor J:
You are on the right track, although the language of CNC is a bit different than woodworking. More like relearning how to mow the lawn! I asked one of my mentors, "Just how much G code do I have to learn?" He responded "Oh about enough to order a Mexican meal." Took me a bit to understand it, but he was right! You can do some cool stuff with a CNC and it will make many things we only stare at possible.

Click here for higher quality, full size image

Click here for higher quality, full size image

Click here for higher quality, full size image

From contributor E:
I have been kicking the whole thing around for a long time, and I had pretty much decided to just rely on other shops to do the cutting for me and keep my overhead lower. But then I found a used router on clist that was reasonably close and reasonably priced, so I took the plunge. Having run KCD since the 90's it would have been a very easy transition to upgrade to their CNC version, but the 12 grand is tough to justify in a small custom shop, as is 80-150 thousand for a router.

As I was poking around trying to justify the cost of KCD, I ran across a Laguna video that featured Mozaik. I did some checking and found out that while it is a new software, the guys that are developing it developed Cabinetvision in the 80's before selling it to Planit, and then they spent time selling CNC routers, so they have a good track record.

For 125 dollars a month you can rent a seat for the full CNC version of the software and they can make a post processor (the program that tells your machine what their software wants it to do). They made a post for my used machine (Shop Sabre) and have been there every step of the way getting the software to work on the machine, and now I can cut all the cabinet parts, cut dovetail drawers, arched valances, shape any cabinet part, and do so many things that I could never have done with KCD, and I am just scratching the surface of the software.

If I had known about Mozaik before I purchased the used machine, I would buy a new machine like the Camaster, Shop Sabre, Laguna, etc. and have them load the software and have the machine ready to go on day one, but then again I have learned so much in the past month about how the machine works that it will certainly help me down the road.

Be ready to spend a lot of nights and weekends learning and understanding the machine and getting it to do what you want it to do. Once you invest the time then it can certainly be an asset to the company.

From contributor L:
With regard to the software that comes packaged with machinery, it's important to keep in mind the portability of your product definitions, whether they're in a drawing, a spreadsheet or a g-code program. The least desirable option is programming in the native language of any one machine. A day will come when you will want/need to run that part/nest on a different machine and if you've written the program directly in one machine's language, it's a near certainty that this won't run on any other branded machine. This is the equivalent of creating a document that can only print on one brand of printer, or one specific printer. The portability of a document that can be printed anywhere, and edited easily to create a new document, is the strength of electronic document files, and it's no different for part information. Choosing a software product (or set of products, depending on your needs) that gives you this flexibility is the first step in the long-range health of your business when it comes to working with CNC. In fact, you may discover that you don't need to invest in the hardware at all if you can get other shops with machinery to fabricate your parts - a relatively easy thing to do if your parts information is easily importable.

From contributor M:
The best advice I was ever given years ago was not to "go into" business, but "grow into" business. Every business owner here, while similar in nature, has a unique business model.

All machinery salesmen, software salesmen, tooling salesman, etc. are oriented on presenting their product as the best for your application. They validate any additional costs by showcasing their features and how they can potentially give you an advantage.

I do believe you have options that include not purchasing software if you purchase a machine that has native software capable of parametric programing, and a pod and rail CNC fits your specific business model. In my opinion, g-code portability is useful if you know you're going to buy several machines with different controls, or know you are going to upgrade machines every few years? Outsourcing parts can be done, but it comes with its own curve - quality, transport of goods, warranty issues to name a few. Works great until there is a problem.

None of my customers with under 10~15 employees, couple million in sales do that. All are running CNCs anywhere from 8 to 16 years old! All made it through the recession alive! Some have transitioned into very nice software packages - they grew into it. Some program in the native control environment parametrically and focus on variance reduction within their internal design process. All of these shops flow work through their shop at very high velocities. In fact, it is most of their competitive advantage!

All of the cam packages mentioned are functional, and do a great job, but in a lot of smaller business models, you can start out with the onboard software and make money, good money.

From contributor V:
I agree with contributor J regarding Mozaik (for cabinet design). They rent the software for around $100 per month. By contrast the yearly maintenance for Cabinetvision is $1500, after you spend 20-30K upfront for the software. That money would be better spent on the machine itself.

To answer the original question, virtually all CNC router software programs output files in G code. Virtually all brands of CNC router accept G code. The software company will need to provide a post processor based on which brand and model machine you purchase.

From contributor L:
With regards to contributor M's response, I wasn't talking about g-code portability, but the portability of the part information itself. Companies, if they're successful, grow, and their needs change. Adding a second CNC machine to the production line is a common event, and often they're of different brands, with different flavors of g-code. Building your programs in a machine-specific program like those that come with certain CNC machines makes the transition to posting to a new machine (whether a replacement or an addition) much more difficult. If the data is in a flexible format, such as DXF, XML, CSV, DWG, etc. then posting to the new machine is relatively simple.

As for virtual machining (and I'm not advocating for it, but mentioning it for consideration), it may not be the right business model for everyone, but it is a viable business model proven by customers of ours doing it successfully and profitably.

I concur that the software has to be taken into account along with the machine when integrating CNC into your production workflow, and should never be treated as an afterthought. Many of the machines are similar enough to be considered interchangeable when it comes to making parts. Choosing the right software that encompasses what you need to do today and gives you room to add new capabilities tomorrow is the more important consideration. From our experience, it's a lot easier to change machines than change how you define what you make.

From contributor M:
Several weeks ago I installed a used pod and rail in a company that makes very high end staircases.

This company already had a licensed version of Autocad and they draw most of components. Autocad, as most CAD programs, embeds the utility to save files in .DXF format (dxf is the utility that allows files created in a different brand of CAD to be viewed and or manipulate by other brands).

As with most smaller companies, this company wanted to use CNC technology to have a bottom line impact with minimal investment (sound familiar?). They purchased a used machine (11 years old!), good iron, good price.

The control on this machine allows for opening of .dxf files within the control software and applying tool paths directly to the geometry. The process takes minutes (literally). Essentially this performs the same process that a CAM program does, it's built in. In addition, it's a conversational control which allows them to rapidly and easily program templates and such from field measurements if needed. If you have ever needed something while in the field, you can appreciate this feature. The last thing you want is to wait for someone to "draw it up." This limits them in no way from evolving to a higher end cam package in the future, should their business model dictate.

Historically on average to cut one staircase of treads (all treads slightly different) would take one to one and a half days. Now five to six staircases worth can be cut in less than a day (and that can be improved on if they focus on variance reduction). And they still have uses to discover that will save them time and money while freeing up their most versatile and important resource, their employees. In fact this change in their process has now moved their historical "bottleneck" to another work center - the saw! As you know this is a recurring theme for companies that continually improve, but one they are more than willing to work with!

This company did not need to buy a secondary software to program their machine. They may never have to. In my opinion, they picked the right type of machine/control for what they wanted to improve and their application.

I also have many examples of companies that have no CAD or CAM capabilities, but make money with CNC. It depends on your business model, your needs and how you want to deploy technology. There is no one size fits all, rather a lot of spandex! Contributor J's pictures always amaze me - the quality is great! His work requires additional software. If you are dedicated to making cabinets, stair treads, some light 3D relief, engraving… depending on the control and application, 3rd party software is optional!

If your application fits, you can leverage CNC technology without having to drop a dollar on additional software. And to be totally honest, if you sit back and look at the total process length in time (value stream mapping?) you could end up lengthening the non-value added work (portion of the process where chips are not being made - example - desk work!) which inevitably lengthens the money making portion of the process in added delays. This company already had Autocad, so they integrated what they purchased within their existing capabilities at a level that works for them and is highly impactful. In fact any addition of software in their process will never have the impact the CNC gave them, and if not deployed correctly will add burden to the overall process.

On the flip side, there are instances where you have to have software combinations to be successful. I just think it is misleading for anyone to infer you have to buy specific software to be successful with CNC. That is simply not accurate and I stand firmly behind that statement even though at times it makes me feel like the lone libertarian at a DNC or RNC convention!

From contributor C:
My answer is this. You have to have shop drawings - they are instrumental in flushing out the design and product with the customer. Why not use all that info to go into production and save some time? We happen to have many jobs, and all of them have to have shop submittals, it only makes sense to turn this info into production of casework. This is almost like the accounting software debate. It can print checks, but some still write by hand and then enter it later. Why not capture this time and use it somewhere else?

We have on board software and it's great, but taking design to production is the goal, especially when some of the jobs have 60 to 70 pages of casework and the router can process cabinets, tops, wall parts, and multitudes of other items like window sills. The benefits of software well implemented in the office driving the shop needs to be examined by all, CNC or not.

From contributor M:
I think we somewhat agree, kinda sorta. I think some of this is a factor of scale. I am referencing companies that may not even be able to process a project like you mention (60-70 pages of drawings). I deal with a lot of companies that are 3-5 man bands! Everyone wears multiple hats. Needless to say funds are looked at differently in the small company scale (I have to spend how much more ($5~$20k) to program the router?). They can be deployed in much smaller scale operations, without spending excess funds, and deliver impact, in some instances more impact than if deployed in other applications. Without opening up the spindle utilization debate, I will stop there!