Makerbot Replicator Vs. Ultimaker

[R3VOLUTION]

I'm looking to purchase a 3d printer. But have a problem. I'm going
to get either the Ultimaker or the Makerbot Replicator. But I don't
know which, the Ultimaker from what I understand is much much fast,
we're talking like 6 or 7 times faster than the Makerbot and also
seems to produces a high quality print layers down to 20 microns. Are
these things true, and is that because of the hardware and/ or
software.... or both? I know the Replicator by Makerbot has dual
extruders now and can print on a larger scale, but is slower and
doesn't seem to produce as smooth of prints on the perimeter. Again I
understand that the speed in which these print is a hardware thing, I
think. But is the layer thickness a software or hardware limitation?
Can the Makerbot Replicator produce prints with the same quality as
the Ultimaker just slower?

[ddurant]

> Are these things true, and is that because of the hardware and/ or
> software.... or both

 

Yes, true. I suspect it's capable of layers under 20 microns - the hardware is capable of a little under 1 micron. Ultimaker worked with a local fablab and did several iterations of the machine before it went on sale. People at the lab would test building and using the machine and the Ultimaker people would tweak the design based on their feedback. It may not be perfect but it's a very nice machine.

 

> ..is the layer thickness a software or hardware limitation? 
 

Actually a somewhat tricky question. For all machines, the hardware has a physical limit but telling the software to do 20x the limit works a whole lot better than telling it to do 2x the limit. The software doesn't care - it's just doing math and slicing things up however you tell it to. You could tell the software to generate gcode for layers that are one atom thick and it would happily do it.

 

Based on results I've seen, Ultimaker's probably good down to at least 10 microns, possibly even smaller, which is 10x the physical limit. For the Replicator, we won't know what it can actually do (which will be different than the physical, theoretical limit) until people start printing with them and pushing towards their limits.

[Prescott Ogden]

The ultimaker has a MUCH bigger build area for a similarly sized machine.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

[Brent Crosby]

> . . . The software doesn't care -

> it's just doing math and slicing things up however you tell it to. You
> could tell the software to generate gcode for layers that are one atom
> thick and it would happily do it.

I can definitely come up with objects that fit in a Cupcake's build
area but cannot be sliced by the version of ReplicatorG that I use (a
couple of versions back I think it is 25).

In one case (0.15 mm layer thickness for http://www.thingiverse.com/thing:7957
), it produces and writes to the disk Gcode of 77MB, but it will not
load that Gcode back in to produce an S3G file for the SD card.

This is for RepG software that is a bit behind current. I do not know
if RepG is better now, and I do not have any experience with
ultimaker.

I just want to point out that the software can be an issue.

[Luis E. Rodriguez]

Ultimaker folks have NetFabb as an option and I have seen others user Slic3r. 

You might ask on the comment page for that thing. One user was using repg 24 just fine, albeit 48mins long to slice! Yikes! No mention of SF version. 

I do think as people figure out the hardware, software IS the next step in evolution for these bots. Software interfaces like NetFabb of the software that come with the UP! are pretty damn slick.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

[ddurant]

> Ultimaker folks have NetFabb as an option and I have seen others user Slic3r.

 

It's a RAMPS board and, sorta, gcode is gcode.. The netfabb stuff says "for Ultimaker" but I don't see any reason it wouldn't work on any machine that can do 5D. Netfabb can act as a host too but they've still got a bit of work to do before it's really stable. It also slices 5D, not volumetric 5D, but we're working on that...

 

> I do think as people figure out the hardware, software IS the next step in evolution for these

> bots. Software interfaces like NetFabb of the software that come with the UP! are pretty damn slick.
 

Yes. Definitely.

[TeamTeamUSA]

Software is the next frontier! Both firmware/control and creation/
modeling. This year is gonna be cool!

Go!

=ml=

[BH]

I think that the Ultimaker is far superior for the following reasons;

I bought a completely assembled MakerBot TOM 7 months ago and really
should have bought a kit because I would have saved additional $800
and the "assembled" version has needed multiple dis-assemblies because
of loose fasteners, sloppy assembly, kludgy engineering and a choice
of plywood that does not allow you to tighten the screws an acceptable
amount to keep them from spontaneously loosening. Considering the low
cost of the plywood this is a particularly poor choice of cost
savings. I think I will have to do a complete dis-assembly and
rebuild with Ny-lock nuts / Loctite / super glue to keep the fasteners
fastened for any reasonable length of time. I would hope that
MakerBot makes some vast improvements to their complete bots or there
are going to be tens of thousands of unhappy customers out there.

I am in the final stages of assembling an Ultimaker and the contrast
between the the 2 machines is striking. The Ultimaker frame is a much
higher quality 5 layer plywood that actually allows you to tighten the
screws to a reasonable torque. The design and engineering is thought
out very nicely. For instance in contrast to the MakerBot the
Ultimaker has cable routing, a much larger print area, lower power
consumption, simpler electronics, the ability to print PLA at quicker
speeds and better precision and can be used with NetFabb to simplfy
and speed up toolpath generation.

[Luis E. Rodriguez]

Just show me an Ultimaker print with no strings, there is a reason
they print fast in PLA! And don't get started on bulging Bowden
connectors!

(oh just settle down, I'm just kidding)

Actually we must move past wood cases someday. Metal cases would solve
auto loosening bots, I would think.

Software is key for sure, netfabb or any other fast slicer is welcome.
Like Slic3r.

Luis E. Rodriguez

[Bill Culverhouse]

Luis just wishes he had one.

 

Show me a makerbot printing at .02mm layer height at 300mm/s.

 

:-P

 

My cupcake is good for some stuffs and the ultimaker is good for others.

The right tool for the job. Now quit fighting children.

-b

[ddurant]

> Just show me an Ultimaker print with no strings...

 

Here ya go: http://www.thingiverse.com/image:75846 .

 

You do remember that MBI machines were as stringy as any other machine, possibly more so because they're so slow, less than a year ago, right? A year ago, the MBI hotness was still a DC extruder.

 

> there is a reason they print fast in PLA!

 

Because we can and because ABS stinks up the place?

 

> And don't get started on bulging Bowden connectors!

 

Can we talk about the ABP, too? Or maybe good ol' DC extruders? MK4? MK5 power resistors?

 

The Bowden connectors have been fine for me except the one time I let the machine sit around hot and heat crept up to the tube. Annoying but no big deal - 10 minute fix - and several people are coming out with better, printable cooling ducts that should resolve that issue.. And that wasn't the "bulging bowden" connector issue, which I've seen reported only once - by a newbie who just finished putting together his machine..

[Luis E. Rodriguez]

Maybe once is comes out of beta. ;)

Ha ha, I love the ribbing. 

I honestly haven't seen a well printing one in person and I was at Maker Faire NYC. probably unfair since a lot has changed with software/firmware since then.

http://www.flickr.com/photos/luiserodriguez/6159125015/in/set-72157627714408848

 

Our Hackerspace has a member with one and is ironically taking a break to fix up his cupcake. I do hope to see it kicking once again.

-- 
Luis E. Rodriguez
Sent with Sparrow

[ddurant]

> I honestly haven't seen a well printing one in person and I was at Maker Faire NYC. probably unfair since a lot has changed with software/firmware since then.

 

I think Erik & Martijn were using netfabb, which doesn't have volumetric 5d or reversal yet.. The only other ultimaker I saw there was Bradleys and that seemed to be printing just fine, though I don't know if he was doing stuff that was prone to stringing.

 

> http://www.flickr.com/photos/luiserodriguez/6159125015/in/set-72157627714408848

 

That is pretty bad but, again, this is how everything off a makerbot looked last year until people got reversal sorted out. And, of course, MBI stopped selling people DC extruders. Maybe you could print something of similar size and show us how it compares? If that's a picture of the Ultimaker+, you should only have to cut the model up into 30 pieces or so to print it. Or you could get a Replicator and only have to cut it up into 4 or so..

 

Seriously, aside from that Ultimaker doens't have reversal working well yet, is there anything at all that makes the replicator better than the ultimaker? Ultimaker's set up for dual heads so that doesn't matter. Carting 2-3 steppers around on X/Y means the replicator has zero chance at coming close on the speed. The ultimaker build volume's far bigger. MBI will only talk 'theoretical' z resolution and that's at 2.5 microns - ultimaker's just under 1 micron. MBI does open source but not really 'standard', while ultimaker is just another reprap which will be happier with a wider variety of software.

 

I'm not saying replicator will be a bad machine but I've seen nothing about it that's an obvious plus over ultimaker.

[Luis E. Rodriguez]

Durant Bait, hook, line, and sinker! :)

Very nice print, I remember that one. 

The ABP was a nice a idea but not ideal for many reasons. Oh boy do I remember DC extruders. You guys run a DC motor? <-- Ok that was just me being an ass right there! My uninformed observations of the bowden when I've seen it is a springy hydraulic action going on there. Like a pressured syringe that doesn't retract…enough. Again uninformed.

Ironically isn't PLA an organic compound that may actually be worse for your body!? <-- totally here-say science. My rule is to actually NOT smell any burning plastic! I can't smell abs or pal unless I stick my nose RIGTH up to the extruder. Truth is I LOVE PLA and only tried it for the first time late last year. I switch between often and sometimes mic it in one model. The mk7 has been great for it but some models seem to freak out the nozzle, sounds like that was addressed with a grew nozzle design on the mk8 (the only change)

I do see all extruders converging to what something like the UP! shipped with or current MK7/8, cartridge heaters, all metal. No need for any PTFE/PEEK insulators in extruders, they all leaf/fail eventually it would seem, pro printers don't use them either. No I'm not saying the Up!/mk7 is perfect. And I've only taken apart a Dimension uPrint t peek at its extruder.

Seriously if I could find an official BOM I would start to gather the materials to try it. I see some users that have github BOM's and listings in the groups/UM forums, but really I barely get enough time with the machines I have! 

In reality the comparison between DIY printers is moot as I have seen astonishing prints from a cupcake/mosaic/prusa/thingomatic/ultimaker. In person THE absolute nicest print I have seen was on a homebuilt Prusa. Marllin firmware I believe.

Respect where it's due. What I enjoy about this movement is you can fit in where you want. I will never write one line of code or some such innovation. I can be just a user with the same setup as another and print WAY better because I've turned this knob and that. Mostly thanks to the 5d work and SF40 and above. I'm ready for less knobs.

Ok being a child session over so we can all "get along!"

-- 
Luis E. Rodriguez
Sent with Sparrow

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/9wqP74ymXCMJ.

[Mike Payson]

On Sun, Feb 5, 2012 at 8:42 PM, ddurant <ddur...@gmail.com> wrote:

Seriously, aside from that Ultimaker doens't have reversal working well yet, is there anything at all that makes the replicator better than the ultimaker? Ultimaker's set up for dual heads so that doesn't matter. Carting 2-3 steppers around on X/Y means the replicator has zero chance at coming close on the speed. The ultimaker build volume's far bigger. MBI will only talk 'theoretical' z resolution and that's at 2.5 microns - ultimaker's just under 1 micron. MBI does open source but not really 'standard', while ultimaker is just another reprap which will be happier with a wider variety of software.

Lol, was gonna leave this thread alone, but after this I can't resist...

 

I agree with you all these are benefits of the Ultimaker over the Replicator... But the MendelMax is $650 cheaper in a kit (including shipping to the US), truly open source and a true RepRap-- no laser cutter required to DIY. more easily hackable, has a similar build volume (~215x235x170, so larger in X&Y, a bit shorter in the Z), etc. No one has attached two extruders yet, but it would be easy to do so, and while it is set up by default for a traditional extruder, converting it to a dual Bowden's would be trivial and would result in a substantial increase in X build volume (how hard would it be to convert the Ultimaker to traditional? How much build volume would you lose?).

Seems to me the real question should be: "Why on earth does anyone pay $1700+ for a 3d printer kit?"

http://www.mendelmax.com (store will be live this week)

[ddurant]

> Lol, was gonna leave this thread alone, but after this I can't resist...
 

Fine with me.. The question was on replicator vs ultimaker so I was just responding to that.

 

Somebody would be happy with a MakerGear Prusa, too. It's got the current resolution record at 10 micron, though Clem Taylor may (or may not.. I can't tell) have taken it back to the ultimaker camp last week with https://plus.google.com/u/0/107307302925551280199/posts/dNrpk9S9NzE#107307302925551280199/posts/dNrpk9S9NzE - his comment of "So I effectively have 32,000 layers in this 30mm print" makes me think there's some record there..

 

> how hard would it be to convert the Ultimaker to traditional? How much build volume would you lose?).

 

The top of the print head block is level with the top of the frame so if you could put the drives above the block without having to make the big changes to it, you wouldn't lose anything. Dunno if there are leverage/balance issues there, though.

 

> Seems to me the real question should be: "Why on earth does anyone pay $1700+ for a 3d printer kit?"

 

There's performance and there's appearance. Exposed wires and such are a big turn-off for some people so they avoid the Mendel-style machines.. I bet people selling these sorts of printers would see a nice sales boost just by hiding them

[Bill Culverhouse]

On Mon, Feb 6, 2012 at 12:17 AM, Mike Payson <mikep...@gmail.com> wrote:

how hard would it be to convert the Ultimaker to traditional? How much build volume would you lose?

 

As Dave indicated I don't think you'd lose any print volume, it'd be a matter of strapping the extruder
to the top of the head. The reason why I'm not upset about the Replicator is that I see no functional
difference (from the videos) between Replicator and UM in that regard so I could "convert" my UM
and have what MBI is selling with a larger build area. Of course I'd have to lower print speeds, but
the Marlin firmware most UM's are using has great acceleration control built in so it would be easy
to tune the bot to handle a big heavy extruder head. Is there any word about the firmware for the
Replicator? Does it support acceleration and look-ahead? Better yet it'd be nice if it just supported
Marlin off the bat.

I've been thinking about making a swappable head that can go bowden or direct extruder for the UM.

-b

[ddurant]

> Durant Bait, hook, line, and sinker! :)

 

Good grief.. This isn't "my machine is better than your machine"..

 

> The ABP was a nice a idea but not ideal for many reasons.

 

They've been selling ABPs since MakerFaire 2010. What's the #1 answer here, today, when people post about their ABP problems? It's to take the thing off and just use the HBP. Saying "not ideal" doesn't really sum it up, Luis.

 

> Ironically isn't PLA an organic compound that may actually be worse for your body!? <-- totally here-say science.

 

I've been using Ultimachine PLA which is NatureWorks Ingeo. You can read more about that here, if you want: http://www.natureworksllc.com/the-ingeo-journey.aspx . Nothing in the safety sheet jumped out at me as being overly nasty.

 

> In reality the comparison between DIY printers is moot as I have seen astonishing prints from a cupcake/mosaic/prusa/thingomatic/ultimaker. In person THE absolute nicest print I have seen was on a homebuilt Prusa. Marllin firmware I believe.

 

Lots goes into making a really amazing quality print. All these machines (except maybe a cupcake) are theoretically capable of doing layer heights that are virtually as good as injection-molded stuff. The big difference is the amount of effort it takes to do those types of prints. From what I've seen over the last couple years, sticking with the OPs question of replicator vs ultimaker, the answer is pretty clear.

 

I saw a post in the ultimaker group recently where somebody was talking about experimenting with really thick layer heights and looking for suggestions on how to print at 0.3mm. I couldn't help laughing at how the perception of what's normal has changed so much, so quickly.

[Sean Tu]

On Sat, Feb 4, 2012 at 5:25 PM, BH <billh...@gmail.com> wrote:

...I bought a completely assembled MakerBot TOM 7 months ago and really

should have bought a kit because I would have saved additional $800
and the "assembled" version has needed multiple dis-assemblies because

of loose fasteners, sloppy assembly...

I find this kind of disturbing, the assembled ToMs were supposed to be fully tested and calibrated. Hope MBI puts some thought into quality control for the Replicators. It seems like they're trying to start selling an appliance instead of a kit, and people's expectations for an appliance will be much higher.

[Jetty]

Without entering the "mine is better than yours competition" :-)

There may be a bit of a game changer shortly. I've been working on
getting the Marlin accelerated driver integrated into
the firmware for the ToM / Gen 4 Interface.

I have it to the point where it compiles and accelerates, the driver
is selectable in firmware between "regular ToM stepper driver" and
"Marlin".

No promises until it's completed, but it's looking likely it is
feasible.

[Luis E. Rodriguez]

Can the Makerbot Replicator produce prints with the same quality as

the Ultimaker just slower?

 

Yes, until acelleration comes out, which seems inevitable. 

 

And Jetty is a genius, his firmware for the Gen 4 Interface is awesome! Instead of just talking about it or teasing in a video he just did it! Open Source rules! Release early, release often! 

 

Luis E. Rodriguez

Are you a Maker?

 

 

 

 

 

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

 

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/IzGO0mBPy3EJ.

[MTO]

Going back to the OP's question:

Do you want someone to answer the phone when you call for support? or
are you happy with trolling internet forums? Do you want to spend time
building? Or just unpack and go?

The ultimaker will probably give you better prints once you assemble
it and calibrate (figure a weekend to assemble, a couple of weeks to
calibrate). The replicator will probably be printing pretty good about
30 minutes after you take it out of the box (print-o-matic and stepper
extruders were like night&day compared to the crappy DC extruder in my
experience), although any time you spend calibrating skeinforge will
pay dividends.

I'm not gonna knock MBI, I like what they've done, but I think they're
moving in a less DIY direction. (I am not accusing them of forgetting
where they came from etc, just that they need to change their tune a
little to go mass-market). If you see yourself as something like an
artist who just wants to print things, doesn't want to fiddle, the
replicator is for you.

If you want to stay in the thick of the DIY/rep-rap community, do the
firmware-upgrade-tango, put in endless hours calibrating, leveling,
etc, go with a mendel kit or the ultimaker. (I'm not criticizing! I
like the tango! Waiting on my new R2C2 electronics for an Air Prusa as
we speak, while I print the RP parts on a TOM...)

If you're somewhere in-between, I'd suggest flipping a coin. For
people who are not going to be obsessive about the difference between
0.3mm and 0.1mm layers (aka, the width of a human hair, approx), the
difference won't mean much, and you can do useful things with a poorly
calibrated cupcake.

[Luis E. Rodriguez]

I think MBI is doing great for a start-up with some 7500+ bots in the wild!

 

I don't think it's anti-DIY. It's pro-convenience. They NEVER will have a warranty sticker and they already said they'd assemble/test/disassemble one for you if you wanted. Why would you want that? Don't you want to just print? I like the tango of hardware and software contributions form the community! I run jetty's firmware and plan to switch the Tom to linear bearings...because I can! And I'm sticking the new electronics in when they come out, NO MORE SHIELDS! Hmm, something nice about discreet seperate boards to replace but man do I dislike wires.

 

Not even sure what we are talking about anymore. You can print crap or works of art with ANY machine. Choose what you want to spend, buy anything and fail repeatably until you understand what this knob does and what that knob does. Better yet find a hackerspace or MB user group, don't suffer in silence. :) Nothing beats a real-world user.

Luis E. Rodriguez

Are you a Maker?

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

[ddurant]

>> Can the Makerbot Replicator produce prints with the same quality as the Ultimaker just slower?

> Yes, until acelleration comes out, which seems inevitable.

 

And you say that because.. ?

 

The ultimaker has had acceleration for +6 months now. Why would adding it to replicator make it as fast as an ultimaker when the replicator x/y stage is so much heavier? No, I don't know what it weighs but it's moving 2-3 more stepper motors around and the ultimaker is moving 0 stepper motors around. That math looks pretty easy to me.

 

For the quality part of the question, all MBI is saying is the theoretical limit, which is 2.5x taller than what the ultimaker has. The pictures I've seen look like they're in the .3mm range - are there pictures of higher quality prints that I've missed?

 

Again, not saying it's a bad machine or that people shouldn't buy it or anything like that. If you're comparing it to an ultimaker though, the specs all seem to say that replicator will be slower and have less resolution.

 

None of these machines is perfect and none is the obvious best answer for everybody - folks should do research, figure out what aspects are most important to them and make their own decisions. The hardest part is figuring out what's just marketing and what's real.

[Dan Newman]

On 6 Feb 2012 , at 8:33 AM, Luis E. Rodriguez wrote:

> I think MBI is doing great for a start-up with some 7500+ bots in the wild!
>
> I don't think it's anti-DIY. It's pro-convenience. They NEVER will have a
> warranty sticker

Ummm, when you don't have a warranty here in the States, you then typically
have an implied Warranty of Merchantability. Wikipedia has some articles on
this, should you be interested. And, also keep in mind that it's not
necessarily New York law which applies here: the deal is "consummated" in the state
where the buyer lives and as such state law from the buyer's state tends to
kick in. (MBI made an offer to sell to customer X from New York; but customer
X made the decision to buy consumating the deal from whatever state they were
in when they made the purchase.) As such, even if MBI makes explicit claims
that there is no explicit or implied warranty, that make not fly if the
purchaser is from a state which prohibits that.

**** I'm not a lawyer ****
**** I'm not providing legal advice ****
**** The above could be anywhere from 0 to 100% wrong ****

The above is is my understanding from having to deal with contract law while
running an international company (which I and my partners eventually sold to
Sun Microsystems).

Dan

[Prescott Ogden]

Dan,

I don't think Luis is talking about the existence if a warranty, but rather the ubiquitous "warranty void if removed" stickers. These stickers represent the anti-DIY policies and attitudes of most consumer electronics manufacturers.

Prescott

[Len Trigg]

That trick of avoiding layer steps by "slicing" in a spiral was also done by Adam with his lampshade script (I don't recall how long ago, ages. I think it is on thingiverse) - he also varied the extrusion width according to a mask to give the result an embossed look. At any XY position you will still have layers in Z (which would probably be the resolution number you are after), just that the Z offset varies as you go round the object.

Cheers,
Len.

> --
> You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

> To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/6b66f4EdSr0J.

[ddurant]

> If you want to stay in the thick of the DIY/rep-rap community, do the
> firmware-upgrade-tango, put in endless hours calibrating, leveling,
> etc, go with a mendel kit or the ultimaker. (I'm not criticizing! I
> like the tango! Waiting on my new R2C2 electronics for an Air Prusa as
> we speak, while I print the RP parts on a TOM...)
 

I'm not sure that's really accurate. The guy who did the 0.02mm prints last summer on an Ultimaker was new to 3D printing and had only been using it for a couple months.

 

If a newbie can get that sort of print in that amount of time, I think most people can probably do 15x that resolution fairly easily..

[Luis E. Rodriguez]

I meant faster than they do now, not faster or as fast as the ultimaker. Ulitmaker mechanics and the Bowden setup allow for that speed.

 

I'm not sure what is marketing and what is theory either. Too many variables when a human assembles the kit. I was excited to see a four extruder ultimaker after seeing that unicorn print switch to red prior to launch. I was sure UM was going to be the first to dual extrusion for use in support. I'm waiting to see a print that swtiches to that flexible PLA for hinges or some such feature. If it's available in 1.75 I could do it now.

 

I don't think any Replicators are shipped so we wont see anyone pushing the limits of it just yet. And I guess since they optimized the new nozzle for PLA we could try these insane resolutions.

 

If these bots worked as stated than neither the makerbot or ultimaker groups would have so many posts! :)

Luis E. Rodriguez

Are you a Maker?

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/1AiAPNyCm-MJ.

[Prescott Ogden]

My ultimaker calibration experience was much better than my thing o matic calibration experience. I just built it and it printed wonderfully using recommended SF settings.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

[ddurant]

> I'm not sure what is marketing and what is theory either. Too many variables when a human assembles the kit. I was excited to see a four extruder ultimaker after seeing that unicorn print switch to red prior to launch. I was sure UM was going to be the first to dual extrusion for use in support. I'm waiting to see a print that swtiches to that flexible PLA for hinges or some such feature. If it's available in 1.75 I could do it now.

 

Marketing-ware is sometimes quite difficult to spot!

 

I think the dual extruder stuff (except for having a separate support material) sorta falls into the marketing category, though, again, I'm not trying to put people off from trying it. The software just doesn't make dual-color easy enough for the majority to use it - until we move off of STL and onto AMF and people start publishing files in that format and/or easy editors like sketchup start using it, I suspect it won't really take off. Sounds really cool + hard in reality = marketing.

 

A dual extruder for support is a different story since both SF and netfabb already (in theory) support it. I again am a little suspicious of the ease, though. MBI has been selling PVA for a year now and I'm just now seeing prints that use it. And those are from Tony Buser who's a bit of a heavy hitter. Sorta akin to saying "well, nophead can do it!" which doesn't mean much for the rest of us...

 

> I don't think any Replicators are shipped so we wont see anyone pushing the limits of it just yet

 

Wouldn't it be nice if it was MBI pushing the limits of their machines?

[Mike Payson]

On Mon, Feb 6, 2012 at 5:55 AM, ddurant <ddur...@gmail.com> wrote:

> Lol, was gonna leave this thread alone, but after this I can't resist...
 

Fine with me.. The question was on replicator vs ultimaker so I was just responding to that.

 

Somebody would be happy with a MakerGear Prusa, too. It's got the current resolution record at 10 micron, though Clem Taylor may (or may not.. I can't tell) have taken it back to the ultimaker camp last week with https://plus.google.com/u/0/107307302925551280199/posts/dNrpk9S9NzE#107307302925551280199/posts/dNrpk9S9NzE - his comment of "So I effectively have 32,000 layers in this 30mm print" makes me think there's some record there..

I'm at about 2200 steps per mm on my firmware, so you do the math what the theoretical resolution is. In practice, I can see no reason why the Ultimaker is inherently better at super fine layer heights, with the possible exception of higher quality leadscrews. Of course MendelMax is easy to hack, so you can easily and reasonably cheaply swap super high grade leadscrews into the MM in just a few minutes (for example http://techpaladin.com/store/#ecwid:category=1941327&mode=product&product=8445245). With the rigidity of the MM and quality leadscrews I don't see any reason why you shouldn't be able to match the Ultimaker's Z resolution (personally, I am way to impatient to care).

 

 

> how hard would it be to convert the Ultimaker to traditional? How much build volume would you lose?).

 

The top of the print head block is level with the top of the frame so if you could put the drives above the block without having to make the big changes to it, you wouldn't lose anything. Dunno if there are leverage/balance issues there, though.

Interesting. So not point either way on that one.

> Seems to me the real question should be: "Why on earth does anyone pay $1700+ for a 3d printer kit?"

 

There's performance and there's appearance. Exposed wires and such are a big turn-off for some people so they avoid the Mendel-style machines.. I bet people selling these sorts of printers would see a nice sales boost just by hiding them

The number of visible wires on the MendelMax depends entirely on the care with which it is assembled. If you put some minor effort into hiding the wires, they are not very exposed at all: < http://www.flickr.com/photos/39800333@N05/6828070281/in/pool-1824888@N20/ >. If you ignore the wires to the display, thay could easily be enclosed, there are not many exposed wires at all, and what few remain could be hidden pretty easily (Personally, I enclose all the wires in braided expandable sleeving, which makes the wires look clean even when you cannot hide them entirely-- and this sleeving is included in my kits,). Of course of you look at the back of that machine you can see the wires, but that is true of the Ultimaker as well (and when you look at the electronics on that machine, understand that he is doing some unusual things with his power supplies so his wiring is much more complex than the average MM). In the next revision, I plan on moving the electronics under the table, so that will reduce the visibility even further.

Don't get me wrong, I like the Ultimaker and think it is a great design... I just think that debating the relative merits of those two printers without considering that there are others that are just as good and cost a whole hell of a lot less is sorta burying your heads in the sand. 

BTW, still no photos in the store yet, but <http://store.MendelMax.com> is now live and ready for your order. :-)

Mike aka Maxbots

 

[ddurant]

> I'm at about 2200 steps per mm on my firmware..

 

Isn't that a hardware/electonics thing instead of a software/firmware one?

 

> With the rigidity of the MM and quality leadscrews I don't see any reason why you shouldn't be able to match the Ultimaker's Z resolution (personally, I am way to impatient to care).

 

Impatient or not, people will ask what it can do. Ultimaker and the MakerGear Prusa have set the bar pretty low - if you ignore that bar, you get threads like this one here. Not saying you need some 0.01mm layers but if you can show some prints at under 0.1mm, you'll have a better answer than "I don't see why not!"

 

> The number of visible wires on the MendelMax depends entirely on the care with which it is assembled...

 

I was just sayin'. The forthcoming pictures should show the machine in its Sunday best!

 

>  Of course of you look at the back of that machine you can see the wires, but that is true of the Ultimaker as well (and when you look at the electronics on that machine, understand that he is doing some unusual things with his power supplies so his wiring is much more complex than the average MM)

 

No, not really much for wires behind it. The printhead power/fan wires are wrapped up with the bowden and don't show very much. Random flickr showing the back..

 

http://www.flickr.com/photos/71653514@N04/6818241647/

 

..and you can see a (black) cable duct on the inside front/left corner. There's one on each inside corner (IIRC, my machine's out on loan) to hide/secure the wires for the steppers and endstops.

 

For the electronics, I have on idea.. All I know is that it's it's a RAMPS deal with a laptop-style power brick - I just hooked up the steppers, endstops, fans and heater to the RAMPS board and plugged in the power supply. Done.

 

>  http://www.flickr.com/photos/39800333@N05/6828070281/in/pool-1824888@N20/

 

LOL.. I saw the speaker behind/left of the machine and thought it was a stepper motor.. :)

[Mike Payson]

On Mon, Feb 6, 2012 at 8:21 PM, ddurant <ddur...@gmail.com> wrote:

> I'm at about 2200 steps per mm on my firmware..

 

Isn't that a hardware/electonics thing instead of a software/firmware one?

Yeah, that was badly phrased, but the point remains: The limiting factor on Z resolution is the Z steppers accuracy and drive mode. At 1/16 step mode, I have a theoretical resolution of well under .5 micron. (2266 steps per mm and change to be exact). Unless there is some other factor that I am missing that effects the Z resolution, that seems plenty high enough.

> With the rigidity of the MM and quality leadscrews I don't see any reason why you shouldn't be able to match the Ultimaker's Z resolution (personally, I am way to impatient to care).

 

Impatient or not, people will ask what it can do. Ultimaker and the MakerGear Prusa have set the bar pretty low - if you ignore that bar, you get threads like this one here. Not saying you need some 0.01mm layers but if you can show some prints at under 0.1mm, you'll have a better answer than "I don't see why not!"

This is a fair point. Personally, I am not interested in pushing the quality barriers, That was never my goal. My goal was always to design the best printer possible and let other people push it too it's limits. Because this printer is a direct descendant of the Prusa, if a MG Prusa hold the record at present, than the MM can match it, and will require less tuning (though certainly not no tuning) to match it. I personally can't prove that, but it will happen soon enough, we just need to get the printer in the right hands. In the meantime, I am ok with just saying "I don't see why not!".

 

 

> The number of visible wires on the MendelMax depends entirely on the care with which it is assembled...

 

I was just sayin'. The forthcoming pictures should show the machine in its Sunday best! 

Absolutely. One reason I have not yet posted pictures is I have not  yet had time to post GOOD pictures.

 

>  Of course of you look at the back of that machine you can see the wires, but that is true of the Ultimaker as well (and when you look at the electronics on that machine, understand that he is doing some unusual things with his power supplies so his wiring is much more complex than the average MM)

 

No, not really much for wires behind it. The printhead power/fan wires are wrapped up with the bowden and don't show very much. Random flickr showing the back..

 

http://www.flickr.com/photos/71653514@N04/6818241647/

 

..and you can see a (black) cable duct on the inside front/left corner. There's one on each inside corner (IIRC, my machine's out on loan) to hide/secure the wires for the steppers and endstops.

Ah, ok. Apparently there is a false back that hides the wires? I had seen another photo on flickr that had the wires visible, but it was probably with that back plate removed.

 

[ddurant]

> Ah, ok. Apparently there is a false back that hides the wires? I had seen another photo on flickr that had the wires visible, but it was probably with that back plate removed.

 

Nope, they just go down those black cable ducts to the RAMPS underneath the machine..

 

You can see the duct install here http://wiki.ultimaker.com/Ultimaker_rev.3_assembly:_Frame and the cables getting stuffed into them here http://wiki.ultimaker.com/Ultimaker_rev.3_assembly:_X_%26_Y_motors.

 

It's really a very clean design.

[Mike Payson]

Ah, ok. The picture that I assumed was the back is actually the bottom:  <http://www.flickr.com/photos/31784001@N05/5806171119/>. Makes sense now.

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/XGVrixMQAswJ.

[Tony Buser]

On Feb 6, 8:46 pm, ddurant <dduran...@gmail.com> wrote:
> though. MBI has been selling PVA for a year now and I'm just now seeing
> prints that use it. And those are from Tony Buser who's a bit of a heavy
> hitter. Sorta akin to saying "well, nophead can do it!" which doesn't mean
> much for the rest of us...

I'm certain nophead could do it 1000 times better than me. :) Using
PVA isn't that hard. I'm hoping once more people have dual extruders
and actually start trying to use it, it's going to get a lot easier
shortly thereafter.

[ddurant]

> I'm certain nophead could do it 1000 times better than me. :)

 

Damn that guy!

 

I almost want to buy nophead an Ultimaker just so he can work out the right reversal incantations.. I don't see any technical reason a bowden can't do revesal but it's going to be a pain to figure it all out. He was the one that really worked out how to do it on non-bowden machines, IIRC.

 

> Using
> PVA isn't that hard. I'm hoping once more people have dual extruders
> and actually start trying to use it, it's going to get a lot easier
> shortly thereafter. 
 

I think I've seen one person (you) doing multiple colors and PVA.

 

The point was more that dual extruders aren't really something new and PVA's been on sale for a year. The number of people with machines that could support dual extruders vs the number of people actually using them sounds, to me anyway, like something a little heavy on marketing.

[Far McKon]

I am also not a lawyer (thank god!) but i did want to chime in here.
This is a brain dump reply to many posts, my apologies for the
length.

TL;DR version:
R3VOLUTION, no matter what open source 3D printer you choose, you win.
3d printing in 2012 is 10x since 2010. By 2014 the whole industry will
10x better yet again. The Future Rocks. Either way, you win! And
what do you win? Nifty Plastic Stuff!


Which Bot is Better:
I'm not going to even touch that discussion. Because I spent the last
bunch of months working on The Replicator, I love it. I love it like a
tiny electro-wooden child. I couldn't make an unbiased analysis if I
wanted to. I'll leave that to others.

Acceleration:
We are working on Acceleration on MakerBot starting tomorrow
(literally, since Beta RepG 0031 was just released). I 100% agree it
would have been great to have it 3-4 months ago, but we have had other
priorities. I wish we could have gotten to it sooner, but we choose to
dedicate developers to a new C/C++ skein engine ( Miracle Grue )
rather than working on firmware acceleration.


Miracle Grue:
As long as I'm on the topic. We're prepping to release alpha Miracle
Grue with the month. This is a GPL skein engine. We prioritized that
over acceleration (maybe wrongly? I don't think so, but... ) since it
was the biggest speed problem. It's fast, it's going to be awesome,
and we want to make it amazingly hackable. We don't just want to make
it easy to tweak your hardware, we want to make our software engine(s)
easy to test/innovate/tweak/invent too.


Skein Engines in General:
When we roll in Miracle Grue (MG) into ReplicatorG, we will probably
add other engines, like slic3r, mandoline, to ReplicatorG. Also,
certain company that starts with a 'net' and ends in 'bb' that stopped
by the office to talk last week... (hint hint) Hopefully we will have
news on that front.

Kits:
MakerBot is a kit company! We are becoming an *assembled kit*
company. To bring those less tech savvy folks the wonders of 3d
printing, we need to be an assembled kit company. We still post our
design docs. We still post/share/make build and maintenance
instructions. We're still open source. We're not abandoning the
hackers and inventors. We are making changes to embracing artists,
modelers, and crafters. As an avid Linux user, I see these same 'no
longer a hacker company/product' discussions in Ubuntu and Debian user
groups. They are suffering the same growing pains as they try to be
more user friendly. All I can say, there or here, is "Sorry, but those
changes are the cost of making the future accessible to everyone".

Layer Height (Can The Replicator print the same resolution as
Ultimaker):
I have a test print here from sf47 at 100 micron layer height in ABS
(from the RepG that beta'd today). If you want to test that out,
please do! I'm pretty sure we can get 50 microns in PLA once we get
around to testing it. If you do test sf 47 with tiny layer heights,
please let us know how it goes, or if you had problems. I know it
works in-house, but I look forward to real world verification from you
folks to make sure sf47 can get that thin. Or, reports and
suggestions if it's not that great in the wild. But don't take my
word for it, go try the beta 0031 yourself!

I hope that answers questions/concerns that I might be able to shed
some light on....


hack on,
- Far McKon
Nerd Herder @ MakerBot Developers Nook.

[Ethan Dicks]

On Thu, Feb 9, 2012 at 4:21 PM, Far McKon <farm...@gmail.com> wrote:
> Acceleration:
> We are working on Acceleration on MakerBot starting tomorrow

Hooray!

> ... I wish we could have gotten to it sooner, but we choose to
> dedicate developers to a new C/C++ skein engine ( Miracle Grue )...

>
> Miracle Grue:
> As long as I'm on the topic. We're prepping to release alpha Miracle
> Grue with the month.  This is a GPL skein engine.

Hooray! (I've been hacking C for 60% of my life, so I'm very happy to
get out of an interpreted language and into a compiled language for
skeining).

I look forward to playing with Miracle Grue (and not just because I love Zork)

-ethan

[mathisyo...@yahoo.com]

gotta love those guys. my favorite's still Far posting that he'd
spend his lunchtime disassembling replicators for those who insist on
building their own bot.

On Feb 9, 4:42 pm, Ethan Dicks <ethan.di...@gmail.com> wrote:

[Luis E. Rodriguez]

**cough** **cough** cloud Miracle Grue **cough** **cough** **send to repg button on thingiverse** **cough**

 

Just sayin'

 

Awesome stuff, can't wait to try it out.

Luis E. Rodriguez

Are you a Maker?

[Len Trigg]

On Fri, Feb 10, 2012 at 10:21 AM, Far McKon <farm...@gmail.com> wrote:
> Acceleration:
> We are working on Acceleration on MakerBot starting tomorrow
> (literally, since Beta RepG 0031 was just released).  I 100% agree it
> would have been great to have it 3-4 months ago, but we have had other
> priorities. I wish we could have gotten to it sooner, but we choose to
> dedicate developers to a new C/C++ skein engine ( Miracle Grue )
> rather than working on firmware acceleration.

Any indication whether the acceleration will work with gen3-5d? (I see
that Rob G has been doing some work on acceleration in his github --
is that part of what you are referring to, or an independent piece of
work)

Cheers,
Len.

[Tony Buser]

I'd just like to add here...

It is also already possible to do 100 micron layer height (ie. 0.1 mm
even though it sounds less sexy) in ABS on a Thing-o-Matic and Mk7
extruder using skeinforge 40 see:

http://www.flickr.com/photos/tbuser/6848488317/
and
http://www.flickr.com/photos/tbuser/6848968813/

As a matter of fact, I have had some success (still needs some work)
all the way down to 0.05 (50 micron if you want to get fancy):

http://www.flickr.com/photos/tbuser/6848969177/

PLA is definitely easier to get down to 0.1 and below, see:

http://www.flickr.com/photos/tbuser/6673840319/
and
http://www.flickr.com/photos/tbuser/6673858363/

PLA at 0.075 layer height:

http://www.flickr.com/photos/tbuser/6747655125/

These kinds of results are going to get easier with a Replicator.
Especially once it has acceleration. (something I've been begging for
forever) ;)

You can download my skeinforge setting experiments here (the latest
ones probably have the best quality):

https://github.com/tbuser/DotReplicatorg/tree/master/sf_40_profiles

[Mackeyorama]

This is fantastic news:

>Acceleration:
>We are working on Acceleration on MakerBot starting tomorrow
>(literally, since Beta RepG 0031 was just released). I 100% agree it
>would have been great to have it 3-4 months ago, but we have had other
>priorities. I wish we could have gotten to it sooner, but we choose to
>dedicate developers to a new C/C++ skein engine ( Miracle Grue )
>rather than working on firmware acceleration.

I have been fiddling with SJFW for weeks now and had started writing
my own acceleration TOM firmware. One less thing to do!
Thanks for the detailed update Far McKon. Lets have more of that in
the future! Keep up the great trail blazing work.

[starno]

I liked the comment about being the right tool fit for the job. I
think that goes for all the major low-end printers out there, as they
have mostly taken different approaches to satisfying certain desires
of the consumers.

In my personal opinion, I don't think striving for sub-50 micron layer
heights has merit, not with existing hardware on any DIY 3d printer
currently on the market today. The build platform needs to have a
flatness tolerance much less than that or you'll be limited to
printing hi-res objects with a small base layer footprint.

I'm far more interested in striving for part vs. CAD accuracy while
eliminating post-processing and maintenance, and using my entire build
platform. I also am not willing to part with ABS since as a material
it is far more consistent from batch to batch based on the way it is
produced. It's unfortunate that it requires a heated platform
though. That's not knocking PLA, I love working with that stuff too.

[ddurant]

Heya Nick,

 

> In my personal opinion, I don't think striving for sub-50 micron layer
> heights has merit,

 

Yes and no.. :)

 

In reality, us consumers don't really have much reason to be down there.. I've printed around 0.05mm and there's not a lot of different in hand-feel between that and 0.075mm or so. I can feel the different between 0.1 and 0.05 if I think about it but don't think there's much difference below that. The different between 0.3mm and 0.1mm is very noticable, though.

 

I've been saying for a long time that as you get closer to the design limits, things get harder. If a company only tests a machine down to 0.25mm, people who buy that machine aren't going to have any idea if it can actually go less than that. Will Z wobble be horribly at 0.20mm? Will the hot end snap off? Will the extruders catch fire? Who knows!

 

So the question, at least from me, was what you guys had tested the new machine with. Charles Pax would only say that the theoretical min was ~2.5 microns which, TBH, is a pretty content-free statement.. What would have been far more helpful (or at least not totally unhelpful) would have been to hear "we went through a number of design tweaks and now the machine prints reliably and easily at X layer heights and Y mm/s. Less agressive settings, of course, also work quite well."

 

That right there could have shut down most of this thread quicker than Clothbot in a grumpy mood. That's something that people wondering how Replicator compares to other machines can use.

 

What we got instead was.. no answer. We see a video of Bre with prints from the Replicator but I can see, quite easily, the separate layers of a moving object from a video from several feet away. That's less 'quality' (note quotes!) than I had on my Cupcake/MK4 a year & a half ago and far less than some current machines. Shrug & meh.

 

I'd be happier if you all were pushing the limits then talking about it.. If you say it's new and awesome and I ask why, more than "because it's SO awesome!" would be nice.

 

Anyway.. TY, to you and to Far, for stopping by and providing some answers. Always good to hear from you guys (even when I'm bitching at you :) ).

[ddurant]

> It is also already possible to do 100 micron layer height (ie. 0.1 mm
> even though it sounds less sexy) in ABS on a Thing-o-Matic and Mk7
> extruder using skeinforge 40 see:
>
> http://www.flickr.com/photos/tbuser/6848488317/

Flickr says this one's 19 months old: http://www.flickr.com/photos/52118225@N08/5391552573/

 

:P

[ddurant]

>  19 months

 

*13 months (sorry for spam.. don't want Luis jumping on me for overstating! :) )

[Luis E. Rodriguez]

:P 

Virtual nuggie!

-- 
Luis E. Rodriguez
Sent with Robin

On Friday, February 10, 2012 at 8:33 PM, ddurant wrote:

>  19 months

 

*13 months (sorry for spam.. don't want Luis jumping on me for overstating! :) )

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/-a4lJK1926kJ.

[Aljosa]

I figured id come and chime in here as well and reply to a few things
since iv been working on a bunch of this stuff for0 the past several
months as well. I think Far and Nick hit a bunch of the major points
on the head really well.

First things first regarding resolution and accuracy.

Iv spent the last several weeks working with SF47 some of which has
included low layer height stuff. I have easily pushed the machine down
to .1 with pretty much no problems, but this isn't really anything new
as people have done this on TOM's, and several other machines in the
past. In the past few days I have taken that one step further and gone
down to 50microns with pretty good success. Although at these layer
heights several factors come into play that make it much more
challenging. This was all in a pretty short time frame too so there is
still plenty of room for improvement. Moving on from SF47 is my goal
while SF35 was nice there is quite a bit better about 47. However we
test and retest things to make sure they absolutely work. So you will
never find MBI just jump on something before its been adequately
tested. Also as Far mentioned theirs a certain other slicer that I'm
actually looking into.

As far as the lower layer height thing is concerned its great and all
that we can push the limits of these machines and get pretty looking
objects. However I'm in agreement with Starno. I would much rather
strive to achieve good printed part to CAD model accuracy than I would
a 50micron print that's not accurate. That resolution doesn't do
anything for me if my part is still inaccurate when compared to my CAD
file.

Speed and Acceleration

This is also an area that I have pretty extensively pushed the
machine. With the Replicator™ I can pretty confidently say that the
mechanical structure is sound, and can move fast despite looking like
it can't because were moving 3 steppers around on top of the gantry (2
for the single extruder model). My work has revolved around the dual
extruder model as anything that setup can handle its pretty certain
the single extruder can as well. Iv tested at several different speeds
all ranging from 40mm/sec to 100mm/sec, and I have had successful
prints at everyone of of those speeds.I'm actually fairly certain that
some of the stuff being printed at CES was being done at 70mm/sec if I
recall correct.

The machine itself is going to perform even better once we put
acceleration on it. Iv been bugging software about that for awhile now
too. But like Far said priorities were made and something else
important took priority. The slicer is the next big step in this ever
evolving 3d printer world in my opinion. There is an enormous amount
that can be done better, and that's why I too believe the right
decision was made when prioritizing.

These seem to be the main things people are asking about I hope some
of this answers a few of peoples questions.

[CreativeDaemon]

Thought I'd chime in on this one. Even though the thread is old it still comes back as a top result on Google. 

I own both the Replicator2 and Ultimaker, having well over 1000 hours of printing on the Replicator2 and over 500 hours on the Ultimaker, I think I can give an accurate answer :)

Both have their merits but when it comes down to it, you're picking:

Replicator 2 for reliability, easier to prepare prints

Ultimaker for speed, potential for higher quality prints, and bragging rights.

The Ultimaker *can* be faster than the Replicator 2. Unfortunately for me I've had to crank back the speed because of faulty prints when printing at high speeds.

My main gripe with the Ultimaker is the infills, which may be a software issue but I've experienced the problem with both Cura and Netfabb. It does an absolute crap job with them. They rarely ever lay down nicely which leads to funky top layers. This could just be incompetence on my part, but I doubt it. I've tried tweaking this thing numerous times with the same results.

The Replicator 2, with the latest extruder and a glass build plate, is a fairly reliable machine. Had some issues with their thermocouple but they seems to have fixed it with their latest revision(first was using solid wire, latest uses stranded).

So sure, you *can* print faster and at a higher resolution with the Ultimaker. At a cost of longer setup, tweaking, and headaches.

For me, it comes down to what types of prints will you be doing.

Large prints with infills? Replicator 2. The Ultimaker doesn't handle these well in my experience.

Vase-style prints? Ultimaker. Just because you can crank up the speed with these style of prints and the shell printing is amazing.

Small, solid parts? Either really. I typically choose the Replicator2 for these just because the Replicator fights with me less.

Artist looking for the absolute best quality prints? Ultimaker.

Running a business? You have no business buying an Ultimaker. You're asking for headaches.

Both are a nightmare to get parts for. It's incredibly annoying having your machine go down due to a failed part and have to wait 2-3 weeks for a replacement. Makerbot only makes a few of the replacement parts available for order, so replacing something like the loves-to-break thermocouple requires a support ticket.

Both will need their acrylic build plates replaced if you print large parts. Replacing the Replicators is fairly expensive in the $60-$90 range but it's a drop in replacement. Replacing the Ultimaker's is a little more complex. I've switched my Ultimakers build plate out with a RepRap style MDF+Borosilicate glass bed and it's working great.

All in all, both are good machines. The Ultimaker is better if you like tweaking things and like the bragging rights. The Replicator is better if you require reliability.

The wooden parts on the Ultimaker tend to drive me nuts as well. They require more maintenance and although I'm sure the Ultimaker people will scream "it's lighter". Forget that, this isn't a sports car, it's a printer. The only reason I can think to use these cheap wooden parts is to keep costs down and at ~$2250 there is no excuse for that.

My business has been growing and we needed to purchase a third printer. We decided on another Replicator 2. For us, the reliability factor and infill quality(we do large prints, typically filling the build volume) made it a no-brainer.

On Monday, January 30, 2012 3:44:02 PM UTC-5, R3VOLUTION wrote:

I'm looking to purchase a 3d printer.  But have a problem.  I'm going
to get either the Ultimaker or the Makerbot Replicator.  But I don't
know which, the Ultimaker from what I understand is much much fast,
we're talking like 6 or 7 times faster than the Makerbot and also
seems to produces a high quality print layers down to 20 microns.  Are
these things true, and is that because of the hardware and/ or
software.... or both?  I know the Replicator by Makerbot has dual
extruders now and can print on a larger scale, but is slower and
doesn't seem to produce as smooth of prints on the perimeter.  Again I
understand that the speed in which these print is a hardware thing, I
think.  But is the layer thickness a software or hardware limitation?

[Jetguy]

The infill is related to the fact the Ultimaker is a bowden extruder.

The problem is, there is no way a bowden extruder can control the pressure in the nozzle due to a number of factors all bowden related.

For example, you constantly fight the filament diamter VS tube diameter and since the arch varies, so does the filament in contact with the tube and thus a huge variation in distance (all relative, it might only be a mm but it matters more than you think).

Then the Tube also stretches slighly in that long of a length and there are all kinds of places for even more slop. It all adds up

 

So yes, I too liked my Ultimaker when I first got it, but honestly, due to stringing, ooze, and uneven fills, I all but stopped using it and only use direct drive extruder bots (My Ultimate T-O-M, or Replicator 1 XL dual).

Kinda the same reason the Rostock Max sits idle right now, bowden plus some sloppy arms/joints = less than great or accurate prints.

 

FWIW, at Maker Faire, I will be demonstrating the Ulti-Replicator.

This design merges the the 2 style bots so it's not a traditional H gantry bot, but it's also not a bowden.

It runs Sailfish since it uses MakerBot electronics and MakerBot extruders, but it has all the weight reduction of the Ultimaker system while also keeping the X-Y motors attached to the frame, not the gantry.

That means no stupid X motor cable problems, and yes, I realize using the correct flex rated wire is an option, but it's better to just never have the motor mass moving on the axis in the first place.

[Minestorm]

https://www.ultimaker.com/p/our-printers/ultimaker-2

 

Good specs and similar price to the Rep 2

[Wingcommander whpthomas]

https://www.ultimaker.com/p/our-printers/ultimaker-2

 

Good specs and similar price to the Rep 2

Still has the bowden extruder... have they improved it at all?

[funmakerbart]

Henri! Give some credits to our Dutch friends!  Regarding speed is Ultimaker=Your design bike and Makerbot=Baby buggy. I'm sure that's because of the light printhead only possible with a Bowden setup (for now)

BTW: it seems to get a Jetguy approval ;-)

https://groups.google.com/d/msg/3dprintertipstricksreviews/Y9BgqsHYhho/I0qNE5PX0jAJ

The direct drive for the Bowden looks very sleek IMO.

https://www.ultimaker.com/p/our-printers/ultimaker-2

Bart