Print Orientation?

@hunter_yaw
No update?

Apologize for the delay - update is as follows

This week we are working on two things:

1) Passing oriented parts to External Manufacturers

2) User Set Orientation for SLS

The first is required in order for us to have consistent orientation across the platform (i.e. for parts both printed in our factories and by External Manufacturers).

The second is hopefully a feature some of you will be excited to see. We originally set out to provide consistent orientation, as you will see in my first post in this thread. Along the way we were able to identify a solution to allow us to capture orientation as set by the designer for Small & Meidum SLS (Strong & Flexible Family) parts.

It will work as follows: The model file will be uploaded into a viewer which can be accessed in Model Edit. The Designer can spin the model by 90 degree increments on the X and Z axes. We have also built a 'Stepping Render' in ShapeJS which will change based on the orientation you select to allow to you see the impact on stepping that a given orientation will have. When you're happy you hit 'Save' and we will print your model in the orientation you select going forward until you ask us to do otherwise.

I have attached a screenshot of the test version we used - it currently has no design work whatsoever so will look very different when it is done but I wanted to show you guys some of the work we are doing.





Any questions please post here.

Thanks!

Hunter

OUTSTANDING!!!

Awesome work, can't wait to reorder my product in a chosen orientation.

Seeing a Visualization of the steppin will go a long way to making it easier for designers to make a choice regarding orientation, how accurate do you intent to make the visualization?

"Seeing a Visualization of the steppin will go a long way to making it easier for designers to make a choice regarding orientation, how accurate do you intent to make the visualization?"

Of course, this only addresses one part of the problem once you get past the the S&F family of SLS parts.

Consider the Frosted Detail acrylics...

If a designer has no idea what his/ her orientation choice does to stepping, then they certainly have NO idea where their choice will result in support material being required. In these materials, that is often a more serious concern. How will you address that?

Why must experienced designers wait until these "tools" - no, more correctly, gimmicks, are created before we can specify orientation in our FUD and FXD parts?

Enough is enough, guys. We've waited far too long already, when choice of orientation is a standard feature at pretty much every other print agency.

It isn't rocket science. If you delay too long, your internal process concerns will lose a huge amount of business.

Steve Hunter

@Fungi @Synio

Great to hear

@CybranKNight

Good to hear and great question - what we are thinking now is that we will actually exaggerate the stepping for larger models to make it easier to see and have it be entirely realistic - i.e. reflective of the current layer height for SLS internal production which currently is around 0.12 mm.

For context in the user-set orientation tool you won't be able to move the camera but you will be able to zoom.

@sbhunterca
This is exactly why we aren't ready to offer user set orientation for FUD/FXD right now. For small and medium parts in SLS orientation has little to no effect on price. For FUD/FXD and any other process which requires support materials the price can increase exponentially (together with the likelihood of a tray crash) as a result of even small changes in orientation.

As a result, building a user set orientation tool for FUD/FXD would be much more complicated than for XSF because we would need to

1) Build a price calculator that would enable the user to see the impact of a given orientation on the cost of the model.

2) To provide context for that price we would need to build a visualization tool to show which parts of which surfaces would require support material for a given orientation.

We looked into how this might be done and established that the amount of time and effort involved would be an order of magnitude greater than that required to allow user-set orientation for SLS. It may not be rocket science but it certainly isn't straight forward.

There is also another substantial consideration here in that our beginning to test alternatives to FUD as it is currently produced has led us to believe that in the near future we may move to entirely different printing methods for Ultra Detail parts which would mean that any work done to enable user set orientation for the current methods would immediately become obsolete.

Having said that - if and when we do build a user set orientation tool for FUD/FXD then every line of code that has been written for the SLS orientation tool will also be implemented for the FUD/FXD tool - the difference being that there will be a lot of work done on top of that existing code to allow for price calculation and surface visualization.

I completely understand your frustration at seeing progress for one material and not having the same opportunities for FUD/FXD - but I assure you that all the work we are doing now for SLS user-set orientation is not being done at the expense of FUD/FXD, but rather will be used to enable progress on FUD/FXD orientation in the future.

I also appreciate your feedback on this - the more we hear from the community on the need for user-set orientation for FUD/FXD the higher we will be able to prioritize it.

Then let me reiterate that FUD/FXD is the material I have in mind when I design the vast majority of my products. There are several that aren't uploaded simply because there's no way to ensure the model won't be a $200 hunk of distorted plastic. There are still others I've collaborated with other designers on that I'm not willing to put up the money just to roll the dice and get junk. And there are others whose models I've favorited but quite frankly I know can't be executed properly yet. And on and on.

I can't tell you the pent up demand for accurate models, so long as they are well executed by the printer. In the model railroad hobby alone we've seen the disappearance of the specialty model makers who ran limited run models in brass for hundreds if not thousands of dollars per model. They have become noncompetitive in today's market due to the emergence of high quality injection molded and multimedia models but so far there is nothing to replace them. Where once the more common brass models supported the esoteric brass models now the common models have been done better in plastic. So there exists a void where the unusual, one-off or regional models just aren't being produced anymore. 3D printing clearly has the potential to fill this void, but not at the inconsistent quality we've come to expect. I know Steve and I and others are working on the CAD to realize many of these models, but again, until the consistency is there it's difficult to pull the trigger and buy unless we know exactly what we're getting. It's especially difficult to ask the customer to buy when the results could be awful, like never buy this junk again awful.

So again, the sooner there is a solution to printing FUD/FXD with predictable and consistent results the better. I know several designers want direct control over the orientation, but I just want control over the quality. It probably sounds like the same thing, but the difference is I don't care how you orient my parts as long as they come out the way I want. I'm content to leave it to you guys to figure it out. Just make it happen.

Just to clarify - when you say you will not be able to offer "user-set orientation" for FUD/FXD, does this also preclude "consistent orientation" after the initial print (which would also require some means to store and transmit a fixed orientation matrix across the supply chain, but would not introduce the unpredictable production costs) ?

@mkroeker

So this is where I believe I have some good news for the FUD/FXD community. In building a solution to allow us to offer consistent orientation for the XSF family of materials we have found a way to extend that solution to internally produced FUD/FXD parts. What this means is that once we have printed a part in a specific orientation we will continue to print it in that same orientation going forward.

We are currently in the process of working with External Production Partners to extend this solution to parts produced externally. Because of the high number of massively-multi-part models in FUD/FXD and the complex pricing structure this is a little trickier than for XSF but we are working hard with each and every External Production Partner to find a solution to enable consistent orientation for FUD/FXD.

I don't want to jump the gun by confirming that we will be able to offer full consistent orientation for FUD/FXD while we are still in negotiations, but I can promise that we are focused on this right now and should have a pretty good sense of whether and if so when we will be able to confirm consistent orientation for FUD/FXD by the end of next week.

Could you confirm this will NOT lead to a big price hike on frosted materials (or in the adoption of another convoluted/complicated way of calculating price)? Thanks !

@fx

Yes - I confirm that if we're able to get consistent orientation running for FUD/FXD it will have no impact on FUD/FXD pricing whatsoever.

180 turn in 24 hours?

@stannum

Nope

The pricing won't change for Consistent Orientation which I hope we will be able to do for FUD/FXD.

What I was referring to in the post above where I mentioned the challenges of cost impact and visualization was User Set Orientation.

The difference is this:

For consistent orientation we will guarantee that once we have printed a part in a given orientation (which our team will select) - we will always print it in the same orientation going forward.

User set orientation (which we are planning to release soon for Small & Medium parts in XSF) is allowing the Designer to determine in increments of 90 degrees on the X and Z axes what orientation a part is printed in.

Apologize if that wasn't clear - please let me know if you have any other questions.

People were asking about "price increase" and "user set orientation" (or "quality based in designer input"). Some people will be fine with (re)prints as set by SW, but in other cases it seems the "quality orientation" will have to be other; and from text above it seems to be undefined about price, but implied to be pricier for some materials.

This is excellent news! (Although the fact it hasn't been the norm from the start still beggars belief...)

Can I assume this new team-chosen consistent orientation will include any alterations that come from the usual process of test prints and complaints/dialog with Customer Service where the assigned orientation was rubbish to start? I'm happy to take the tech's suggestion, as long as it can be reassessed as necessary in light of test-print experience before the item gets released to the wider world for sale! I think many of us complainers here would be very happy with that.

@stannum

I don't entirely follow this but I can be very clear about the following - we will not charge a cent more for User Set orientation for Small & Medium parts in SLS and will not charge a cent more for Consistent Orientation when it is launched in SLS, nor will we charge more if we are able to extend it to FUD/FXD.

If in the future we do develop a tool for User Set orientation for FUD/FXD then there will be a dynamic pricing element which will change based on the orientation you have chosen, taking into consideration the impact on the amount of support material and the likelihood of a printer crash (which increases exponentially together with the Z height). However - the price for a given orientation would be shown to you in the orientation setting tool so it would be your decision to choose a specific orientation and pay the difference if it is more expensive than the orientation we would have used.

However - for reasons I outlined above - we are not currently planning to build a User Set orientation tool for FUD/FXD.

So in the man time - no element of either Consistent for S/M XSF parts and possibly FUD/FXD or User Set orientation for S/M XSF parts will lead to any increase in price whatsoever.

@barkingdigger

Yes you can - the current process will remain the same. The only difference will be that we will be guaranteeing consistency for XSF Small and Medium parts (and possibly for FUD/FXD) unless you yourself choose to change the orientation using the User Set orientation tool, in which case we will guarantee that the orientation you select will be consistent going forward (unless and until you decide to change it again ).

@Hunter-yaw - that's music to my ears! (or to my eyes? It's on a screen...)

@MitchellJetten - Is this a case of "The Lord giveth, the Lord taketh away"? While I appreciate the idea of re-evaluating a revised model, why exactly must each uploaded iteration asutomatically reset the orientation to square one? Surely there should be a presumption of retained orientation on any re-uploaded model, but with it being checked for best-fit by a tech the first time it gets printed? After all if I design a box and go through all the negotiations to get it oriented "detail-side up" for best quality results, and I then need to tweak one small detail, I will still want it oriented the same way unless a SW tech can give me a very good reason why it would be better otherwise. To blindly throw away any best-orientation effort already invested seems a bit retrograde to me. Please tell me I'm misunderstanding your statement, and that good old common sense will prevail.

@barkingdigger - this will probably depend on where the orientation is stored, which at this time may not even be certain yet. For production, I imagine they would want it to be in the file so that it cannot get lost or mixed up (provided that is even possible given the variety of supported formats, unless there is already a hidden copy of the model stored in a single printer-specific format). For keeping orientation across generations of a model, having it stored separately would probably be preferred, but may have other drawbacks.