How many people worldwide use Marlin? 10,000? 100,000? 1,000,000? Is there any data or credible estimates?
I don’t doubt that it’s over 10,000. In github alone, there are >2000 forks, >1000 people who starred the project, etc. Shit, every Ultimaker ever sold comes with a copy of Marlin and Ultimaker was already to the point of selling >5,000 machines/year back in 2013.
So, >100,000 is likely.
100,000 sounds like the right order of magnitude.
I ask because a lot of people say Marlin is the widest-used open source 3DP firmware, but I’m pretty sure that’s actually Sailfish. Every Makerbot Replicator 1/2/2x, plus most Cupcakes and ToMs, which makes about 300,000 printers I believe. Plus every Mightyboard RevE clone (FlashForge, Wanhao, Mbot, Weistek, ZYYX, lots of others not common in the US) which at minimum doubles the number. So it’s in the 500,000-1,000,000 range for Sailfish and Makerbot’s “stock” firmware version of Sailfish. But that’s relatively easy to figure out because we have good market reporting on purchased printers. Marlin’s home-brew population is hard to measure. But the pre-built Marlin-based printer sales numbers are tiny in comparison.
@Ryan_Carlyle , IIRC, Sailfish is not the default firmware for the Makerbots/Clones/etc. They come with https://github.com/makerbot/MightyBoardFirmware
Not Sailfish.
@ThantiK Mightyboard Firmware is just a fork of Sailfish that is maintained by Makerbot. They contain the same core motion planner and largely the same features. Makerbot officially supports use of “real” Sailfish on their machines because they’re functionally identical.
Also, even if you exclude machines running Makerbot’s fork of Sailfish, the “actual Sailfish” machines still number in the hundreds of thousands to all appearances.
I see your logic, but I can’t see there being more sailfish machines than marlin machines. I consult on 3D printers, so I actively seek out any place where I will see them in the wild. Even if the machine is not familiar to me, it’s easy to spot an x3g machine because they all have an LCD with a 5-button control interface. Marlin’s LCD interface is based around a rotary encoder, so just the briefest of glances will tell me which firmware it is using. Machines using proprietary/homegrown controllers (such as 5th gen makerbots, Reprappro Ormerod, Tiko, etc.) are few enough and have strong enough branding that they’re easily identified as such, and other firmwares like smoothie/teacup/etc. are virtually nonexistent among printers/kits being sold today (Repetier firmware, which is loosely based on Marlin, is probably second place among non-x3g machines).
When I see or hear about printers that people are actually using in the wild, Makerbots and their clones probably make up a plurality (even that may not be the case, because a couple of the local companies are over-represented), but they’re FAR from a majority. It’s like how I see a lot of Priuses (Prii? Priora?) on the road, but it doesn’t compare to the number of all other cars.
@Whosa_whatsis The way I interpret that observation is that Makerbots and clones are so common (and, you know, heavy due to steel frames) that nobody bothers bringing them to MakerFaire type events. There’s also a meaningful stigma in the maker community due to the Takerbot fiasco(s). But it seems like everybody who builds a Mendel or Mini Kossel wants to show it off. It’s just a different and less vocal/visible crowd buying Sailfish bots than Marlin bots.
The most recent published sales metrics we know about say that Makerbot has sold ~80,000 bots. Given the general flop of the 5th gen line, we know the gross majority of those are Mightyboard-based bots. There are also huge numbers of clones being sold in Asia that aren’t appearing on 3dHubs or in our typical English-language maker communities. I would guess there are five to ten times more Mightyboard-based printers globally than the official Makerbot sales. They’re being made in significant numbers by FlashForge, Wanhao, Qidi, Mbot, CTC, Weistek, ZYYX, and others for export, plus whatever the domestic market in China is. There’s an entire ecosystem of vendors and printer manufacturers in China built around the Mightyboard platform. That’s all built on one flavor or another of Sailfish.
As a sidenote, Sailfish now supports rotary encoders (specifically, the Viki) but rotary encoders are clearly inferior, so don’t expect a stampede of people rushing to ditch their 5-button panels 
Also, Weistek has a Sailfish fork (used by the Ideawerk) that supports touchscreens. And there is some credible suspicion that large fractions of the Dreamer touchscreen firmware (used by FlashForge, Dremel, and possibly others) were originally derived from Sailfish.
@Ryan_Carlyle I agree about Maker Faires, but that’s not what I was talking about. I’m talking about my consulting clients and potential clients, and everyone else I talk to who is using the machines either at home or in a business. Makerbots and their clones are a larger percentage of these machines than of the ones at Maker Faire, but still nowhere near the proportion you’re suggesting.
It’s possible that the people who use them are disproportionately people who won’t go looking for help (and thus people I wouldn’t hear about), but the data I’ve been able to collect indicates that if they skew toward any market segment, it’s people who want a machine that “just works” (rather than those who are interested in messing with the machine and learning how it works, present company excluded of course) and thus who will need MORE help when something does go wrong. Given the problem that I have heard about people having with them, and the difficulty level of fixing them (e.g. PTFE tubes deforming in the clones that use them), I would expect to be seeing a lot more from those people, both locally and in certain online circles (even excluding ones like this community, which may self-select against them) if the proportion really was close to what you estimate it to be. Not saying you’re wrong, just that the data I spend a lot of my time collecting doesn’t support it.
@Whosa_whatsis The only thing I can conclude is that they’re using different channels than you’re looking at. The various Clone Google groups have a ton of activity, far more than the equivalent Makerbot Google group for example. In the communities I frequent, clones are far more common than RepRaps. Tons of people say they buy FlashForges or CTCs for their second or third or fourth printer after deciding their first Makerbot didn’t have good enough value for money.
G+ has a veritable hate-on for all things Makerbot. I can’t believe how overwhelmingly negative and full of bad advice / unanswered questions / Takerbot-bashing it is here.
I think one way to explain the possible lack of traffic that you see is that the lack of hardware and firmware diversity means much less support content is required per user. A few good YouTube videos and Google group posts can cover the maintenance needs of thousands of users who don’t participate in communities directly. These resources exist, and demonstrably get tons of views.
I’m aware of those other channels, and I lurk on a lot of them. You’re probably right about the lower diversity making one-way, one-to-many support options more viable, and thus making the userbase more silent. It explains some of what I see (and what I don’t see). It’s also possible that the chinese clone machines have a much higher market share in chinese (and possibly other non-english-speaking) markets, on which I have almost no data.
Still, I have trouble believing that the average sales volumes of all of the small number of companies selling makerbot/clone machines exceeds the average of all of the Marlin-using manufacturers by enough orders of magnitude to match the difference in number of manufacturers (this is ignoring the people who are not buying a complete machine or kit that includes a controller board that determines their firmware choices, which is now a statistically insignificant group, except at Maker Faires).
I actively seek out actual users of printers in my area in ways that don’t select against these people, so I’m confident that the ratio isn’t skewed the way you suggest in my area.
Actually, a map of the relative geographic distribution would actually be a lot more interesting than the measuring contest of which firmware has more total users.
@Whosa_whatsis That all makes sense to me. I don’t know if Sailfish is 150,000 or 600,000. I also don’t know whether Marlin is 150,000 or 600,000. I think they’re both probably in that general range. Unfortunately I can’t think of any good way to get any kind of rigorous data on the issue. Print job aggregator services like 3D Hubs are decent as an indicator, but they seem to over-represent certain classes of machines due to geography or machine capability.
The thing that’s really fascinating to me is the non-English-language communities. Germany has a strong RepRap movement and favors a lot of technologies that are uncommon in our circles, such as certain permanent printer plates (which I believe are novolacs powdercoated steel sheets). The language barriers seem to cause a lot of “siloization” between communities. Hell, even the Deltabots Google group and Makerbot Users group favor different extruder designs and build surfaces and firmwares and all that. It’s fantasy to expect entirely different language-groups to make the same technology decisions. That self-isolation and clique-forming aspect is really interesting to me.
Yeah, the only thing we can really tell for sure is that the uncertainty in the number of users of each is larger than the difference between the two, and with language barriers and such, it’s unlikely that this uncertainty will ever be resolved.
Fabbaloo just posted some wild-ass guesses relevant to this discussion: http://www.fabbaloo.com/blog/2015/5/31/how-many-3d-printers-are-there-really
I think the response to various kickstarters proves that 20,000 estimate is too low. Or else there’s a huge aspirational demand population of people willing to spend $200-2,000 and then do nothing with it. (Which is possible.)