Introducing 3d Print Orientation For Fud And Fxd

Discussion in 'Official Announcements' started by Andrewsimonthomas, Jul 10, 2017.

  1. stannum
    stannum Well-Known Member
    Different place. SF material, no shops, only manufacturing.
    So low sales, so no reason to improve that side of shops, so low sales, so...
  2. Sparkshot
    Sparkshot Well-Known Member
    It has happened, don't know how many times but it has. I can likely dig out photographs too when on PC next but there is no point. As another has even said it has happened to them too.

    The model of mine you have there looks like the Furness K2. One wonders why you are showing a very out dated mesh that has been improved upon since although I am guessing this is the print history in which case fair enough.

    Either way though you go to some other models and the new tool shows the model on its side. U flip it ANY way and the price is always higher.

    For me personally though I'm not too fussy. I've been modelling for over 20 years so I can get a good finish either way.

    However for customers it does matter as I wanted to get the price down and if possible quality up but it seems either is impossible.

    Even the cheapest is high for what I am selling to what market so it will have to stay.

    Truly stuffed
    Last edited: Jul 15, 2017
  3. railNscale
    railNscale Well-Known Member
    @ Crashtestdummy,

    Let's try to dig through your remarks.

    "CTD: Your right 90% of your customers have no clue about orientation and I bet 99% don't get their underwear in a bunch worrying about it. If you don't want to change the orientation, don't change the orientation. If you do care about the orientation do something about it, easy and you don't need customer service .... "

    > Ofcourse I care about the print orientation. That's why I want the previously used before print setting to be a fixed orientation in order to get the consistentency! The print orientation is nowadays determined by the tech people at Shapeways. They are the experts. Until now SW just deleted the chosen orientation. Now I know that multiple part models are split adnd that the individual parts may be oriented differently.
    Now the deal SW wants to make is to set a default -never printed before, and certainly no good print orientation- for all products and want us to dig through the models to find out how bad these orientations are. The supplied tool does not even allow to set the print orientation the way SW printed them before!

    "CTD: "The main concern is the fact that suddenly changing the printing orientation in a not proper but cheapest orientation without informing us and the customers means that you mislead the customers."
    Nobody suddenly changed anything see the terms and conditions. If you don't like the terms and conditions your back to my four options."

    > Please read the text of the topic starter! It says: "Please note: Until 1.1.2018, if you have not saved a 3D printing orientation for the model, our 3D printing engineers will set an orientation for the model or use the orientation the model was printed before. Starting 1.1.2018, the orientation shown in this tool will always be used."

    So, YES things wil suddenly change! Namely customers will receive crappy parts.

    "CTD: "Misleading or mis-informing customers is the worst thing you can do. Actually the whoel question for setting print orientation started with discontent about the print quality altogether."

    Now your arguing against yourself. Maybe you should make sure the customer reads the terms and conditions or you should lock in the bad positions, The problem is yours not Shapeways. If you feel bad about "Misleading or mis-informing customers is the worst thing you can do." than you could always Inform or properly lead your customers on your page mention the orientations have been change to provide a better prints or like I said offer the cheep orientation and a more expensive "optimized" orientation."

    > Again you simply mis the point. SW is not providing us the required information. SW just throws some tools and unrealistic cheapest print orientation tool directions over the fence. As a result the print quality cannot meet the supposed print quality (see the photos of the printed before products that inform our customers).

    "CTD: I work in manufacturing and have for about 30 years, I have yet to find a customer who was upset the product was better than expected and demanded an inferior product."

    > Interesting sentence. But never raised in the topic.

    "CTD: Giving two options you could hit both the premium market and the budget market and you might even get folks who ordered the cheep orientation reordering the expensive best orientation letting you sell two models to someone who would have only bought one."

    > I would suggest you to check the prices of SW products and compare them to the mass produces parts. That's the comparison our customers make. Then you will learn that the price setting of 3D printed objects is never the budget direction. You use 3D technology to fill a niche. People want a specific item. Besides trying to do both premium and budget market does not even work for multinationals. So, nice theory, but it is nonsense.

    "CTD: ... The end customer doesn't care how its made it is only important that it meets their expectations and arrives in a timely manner at a price they will pay. ....."

    > Bingo! That's were this discussion is about. Now SW is suddenly meddling with the parameters of the production. And they are doing it in such way that you cannot forecast what consequences it will have. You know however it will cost you lots of time and will not likely lead to better products. Again: the provided tool cannot set the print orientation the same way as it does currently. Namely: you cannot chose multiple tool orientation directions. And that's whta SW does now.
    Apart from the enormous work load with little improvement potential (if all) you beforehand know that the product prices will raise. And if you have to re-arrange your designs (modifying for print direction) you get the "first-to-try" and "rejection lottery" as backfire.

    "CTD: As for your variants comment, depending on what your building variants are part of the original product. Boxcars all look basically all the same but when you get down to it they are almost all different depending who bought it when and who made it. Example Pullman-standard made the PS-1 Boxcar out of "standard components" but it could be 40' or 50' in length 6', 7' 8' doors on the side both the roof design and ends changed between the first PS-1 and the last one"

    > Maybe you address this sentence to somebody else. But I'm not doing squary box cars. I'm designing rather not squary products with little commonality on the visual side.

    "CTD: If your selling though the Shapeways store as your sole source of income, how is that shapeways or my problem? ..."

    > Why you come up with this 'IF' sentence is unclear to me. I would suggest you not to add all sorts of sideline discussions. The topic started with "For a long time our community have been asking for a way to control the orientation products 3D printed in Frosted Ultra Detail (FUD). Starting today, you finally can!"
    And how this statement has been mingled with a working procedure that is far from productive and negative price impacts. And that this new feature is sold as "You can" but clearly tells "You have to".
    All in the mname of improvements. Now don't get me wrong: I am a great fan of setting print orientations that suit my designs, but not under these conditions and certainly given the fact that we have invested large figures for our current portfolio.

    In summary:
    1. Yes, we would like to get consistently printed products with a consistent print quality that resembles the products we have been printing before. We do not necessarily want to pre-set the print orientation since we lack the knowledge to define this. SW has the expertise. If SW decides individual print orientations per part in one single model than that's fine. As long as SW goes for quality and keep this pre-set in the future.
    2. We strongly disapprove the default 'cheapest' print orientation, because it does not reflect the print orientation SW chooses when we ordered the products. SW needs to store the used print orientation of the first print. That should be shown in the product information.
    3. SW does not provide sufficient and correct information about the used print orientations. This lack of information makes chosing a proper print orientation impossible.
    4. For shops with a large inventory it is not possible to manually go trough all models. Given the above mentioned items we have no clue what to do.
    5. The print orientation selection tool is not able to replicate the printing orientation of multiple part models. This tool is simply missing the required possibilities.
    6. We noticed that the provided tool is not working at all. It times out and is time consuming.
    7. SW's new pricing philosophy started with cost reductions. It now appears that SW has been misleading the shop owners by comparing apples to pears ('optimized' print orientation vs 'cheapest orientation).
    8. SW's print orientation tool forces us to reconsidering the entire price setting again.
    9. Again SW manages to introduce a disruptive tool that at bests will cost us a lot of time without gaining improvements (see points 1- 8). This will seriously jeopardize product introduction schedules and penalize our sales.

    RAILNSCALE v.o.f.
  4. ETS35
    ETS35 Well-Known Member
    @ Shapeways, having read a bit of the comments here, I notice some people are not happy.
    I know I am very happy we are finally able to fix the orientation.
    It is disappointing that it leads to an increase in price on a number of designs, but the effect on my designs has been less than I feared. But of course increases in price are never nice.

    I've always put quality first, and hated the fact that I had no idea what my customers were getting. I designed sprues for easier handeling by Shapeways, but primarily to prevent their engineers from screwing things up. Orientation has always been a priority to me and from early on everything has been prepared for the moment designers would get to select the orientation. It baffles me that some other designers haven't done so. Uploading files that require different orientations for different parts is something else I really find difficult to understand.

    Still, I do think Shapeways could have done a lot more in the past to educate its designers. That would have allowed them to better understand how they should design for FUD/FXD. It would also have prevented a lot of designers have to modify a lot of their designs in short period of time. The price change should have been announced and explained much, much sooner)

    Any way, I'm relieved we can finally set the orientation. However, I do not think the new tool is perfect just yet. (In itself it works perfectly for me. No glitches, other than the occasional error that it corrected in a few seconds)
    - On some flat designs it is difficult to determine if it is facing up or down. We need a better angle.
    - It would be nice if there is a way to say it the orientation on a specific design has already been fixed by me before.
    - It would be nice if I could set my account setting to immediately fix the orientation to match the files as they are uploaded. I do realize the orientation could be different for different materials, so perhaps we should be able to select this for individual materials. I rarely use anything other than FUD/FXD so this would save me some time.

    @ railNscale, I'm surprised you have trouble determining how a design should best be printed. I've never found it that complicated. Still, some of my models include dozens or even hundreds of different parts which are designed to make best use of the FUD/FXD printing proces. I can image that if you have details on all sides of the design determining the best orientation can be challenging. Is that part of the problem?
  5. railNscale
    railNscale Well-Known Member
    @ ETS35,
    Well in many cases you can determine a 'good' printing orientation. For passenger car bodies the answer would be: print the model with roof up. However we've received models that were probably printed with the nose up (or angled slightly) so the front of the car would have the best print quality. To some extend that could be a better print orientation.
    For other parts, it really is less straight forward. As our models are usually multiple part, I know SW is splitting these parts to get a better filling of the printer tray. I get the impression that often parts coming from one model are printed in a different print orientation. For example sometimes the chassis of a car seems to be printed sideways. But not always.

    Currently our models are uploaded in a way that could be the good orientation. Basically roof-up, and seats of chassis on top side. So two parts next to another.

    I think this would provide a good printing quality. The visible top sides are the smoothest. (Cost price €6,95 in FUD; €7,99 in FXD).

    For this image I've re-uploaded the old car model and I notice that THIS is the position the viewer shows.

    However if I am looking at the same model I've upladed about 3 years ago, SW has reoriented the kit into this:

    This is how probably ALL our car models are positoned into the cheapest orientation. This is NOT a good print orientation. (Cost price €6,31 in FUD; €7,07 in FXD).

    Now I'm sure SW would nowadays not print the car like this. Likely it is either according picture 1, or according the picture below.
    (I composed this picture):
    (The cost price is in between the above pictures) The problem is I cannot select this orientation, because it requires a new upload. Ofcourse one can consider reuploading 800 models of the inventory, resetting prices and throwing them back to the 'print-the-first' time modus and go through the rejection policy hell.

    With the tool I cannot choose multiple orientation directions per part, as SW probably would print it.

    So coming back to the question 'do I know the correct print orientation'? Answer: I guess I know. The uploads were always with this in mind.
    Do I know how models were printed? Answer: I've seen multiple solutions (not consistency problem), but I have an idea how it was done. However I've received 0 information of SW so far. That used to be no problem (because one cannot choose), but now it becomes an issue because SW is basically taking their hands of printing quality starting 1.1.2018.

    The work-around for already uploaded models would be in our case: keep the print orientation with Z-axis up (according picture 1). I noticed the initial view of print orientation I get after uploading is according picture 1 (and that's not the cheapest orientation).

    An alternative for already uploaded and printed before models would be: keep the print orientation according the actual printed conditions (probably something like picture 3 in this case).
    That would be in line with the actual photographs we show next to our models. These photos are key indicators for quality for our customers. As we were always encouraged by SW to do test print kits and to make photos, we have invested in these models. The proposed way of working suddenly whipes away the value of these investments (in the meanwhile we've order about 120 times a bunch of products with one purpose: quality check, processing, photos. None of these items were sold to customers.) These are investments we made in order to get the shop and products on the level we would like and is key to the success we have here.

    It seems that SW is not able to provide us the proper information and to share the successfully used before print orientations. As long as SW cannot provide this info, the proposed 'improvements' are premature. Again: I talk about the already uploaded inventory of our shop. For new uploads sure you can start with the new tool and optimize the work for this.

    If SW wants to improve their quality, I'm happy with that. But the least I expect is that we build upon the proven models we have, and not to start from zero again.

    taz_of_boyds likes this.
  6. railNscale
    railNscale Well-Known Member

    I was wondering how a larger set would look like in the Support material viewer.
    So I picked this product:

    This model contains 9 parts. It was designed in the old days were I thought it would make sense to pack individual parts densily together.

    It contains 9 parts:

    However the support material viewer showed after many many minutes (what a slow tool) just the three biggest parts:
    Her in another view:

    What a rubbish! The tool is not able to show all parts!
    Ofcourse the cheapest print orientation is not good for the main body of the bus.
    What I noticed is that the support tool viewer and calculation tool cannot reorient individual parts. I know SW is doing this all the time when they fill their printers.

    So, unfortunately the tool is no good and it does not reflect the way SW is printing parts. This makes the entire 'more realistic print cost calculation' an incorrect statement.

    It is a pity to see that SW fails to introduce proper tools and tell stories that do not reflect the actual situations.
    I have difficulty understanding this policy.

    Sparkshot likes this.
  7. ETS35
    ETS35 Well-Known Member
    @ railNscale: Thanks for the explanation. I think you are being considerably too pessimistic in what you're assuming.

    I do not see how it is a problem that you cannot orient individual pieces. In your first example the cheapest suggestion seems to consistenly put all you pieces on their right side (compared to your upload). This applies to all parts, even those that are not included in the overview. So it does show how Shapeways will print it (after 1.1.2018).
    This consistent rotation would mean that if you change this back to the top facing upwards, there should be no problem. All the parts in your design will have the same, correct, orientation.

    As for the second example, I know it's an old design, but that should not be a problem because each individual piece will be printed on the tray. The only potential problem I see is having an old design with parts you'd like to use different orientations. If so far you have relied on Shapeways fixing that for you, you are in trouble... Personally I do think it has always been the designer's responsibilty to know how things should be printed, but due to the way Shapeways' used to work you could not control how they did it. And as we all know from misprints: leaving things in the hands of Shapeways' engineers has never been safe. (that's where customer service came in)
    crashtestdummy likes this.
  8. Sonic_Sun
    Sonic_Sun Active Member
    Many do not understand that shapeways had to act like that.

    first. Shapeways has not reckoned with this material that some with huge models which are also still built like a box which then still have to be printed with the open side down. And this costs support material and print space and printing time.

    Secondly. It is a huge effort hundreds of small parts individually (without sprue) to sort and clean. (I myself hafe many small parts offered)

    Everyone wants the best result for their pruduckt, but what when shapeways loses.
    What would have happened if shapeways the material no longer offers, because not economically.

    I will not name any guilty. But I find the models are large and hollow and models with many small items which were not connected with sprue the problem have triggered.
    best wishes
    crashtestdummy likes this.
  9. crashtestdummy
    crashtestdummy Active Member
    In summery I don't think you will be happy with anything anyone says, including me. No point in wasting time as you have made up your mind about what you want to see and hear. Your not going to read it so probably not worth the time to write it. In conclusion, I hope you get what you want as this thread is cutting into my CAD time.
  10. railNscale
    railNscale Well-Known Member
    Maybe I was not clear. There is no problem with the ability to set a print orientation, providing you get the proper information about what choice is best.
    My concern is related to the already uploaded models (it is an inventory with over 800 sets). The systematics of SW is to chose the cheapest orientation by default. As far as I could check these orientations are in about all cases wrong and do not support the intention of improving print quality. These print orientations were (I hope) never used for any prints.
    This means all efforts we did (pre ordering, checking etc) have no value anymore. And this means we have to CORRECT all print orientations in the orientation that will likely be best (namely Z-axis up). We've uploaded our models with in mind an optimal print orientation (based on the global experience we have) by using the Z-axis up = face up direction.
    So SW basically offers an 'improvement' by setting our entire inventory the wrong way. The time to correct this is collosal.
    One notion I made is that some of the printed models of our inventory were printed with different orientation settings per part. And you cannot select multiple directions. This means if you would like to replicate the printed before orientations you have to re-upload models (and then you go in to the 'never printed before' and 'rejection parody'.

    Regarding the second example, I would like to correct one thing. You mention "If so far you have relied on Shapeways fixing that for you, you are in trouble... Personally I do think it has always been the designer's responsibilty to know how things should be printed, ..."
    The reality is the other way around. SW never provided any tool. So yes we had to rely on SW. In whatever direction you would have positioned your products would not mind. SW would always choose the print orientation that suited them best. I agree a designer can design better stuff if he is aware of the processes of course. But again, who is providing this information to you? I've been many times in Eindhoven and had a lot of talks with the SW-people and seen quite some printed objects so far. The choice between print quality vs. optimized price is not that straight forward.
    A simple rule is that if your most viewed surface facing up would often be the best for print quality (Z-axis up). This we did. And due to the transition as proposed by SW we are in trouble.

    @sonic Sun,
    Sprueing small parts together makes sense. Sprueing all parts together (small and big) is not a very good idea for the following reasons:
    1. Larger sprues break easily
    2. The filling of the printer becomes far less economical if you receive connected parts.
    3. The way SW does place parts in the printer is not based on how you deliver parts. An algorithm defines the position and orientation. The technicians will take a look and correct it where necessary, but the filling of the tray is very very different to what you and I would guess. Ofcourse with the a fixed preset print orientation the repositioning possibilities will be limited. That will cost some machine space too.

    Regarding the economical rationale behind the new FUD/FXD print orientation and price settings, I do not know anything. It is pure speculative.

    The whole idea of the FUD/FXD print quality improvement ideas started with a more print consistency and/or ability to pre-set the print orientation fixed.
    The thing is that while this becomes possible, SW proposed to do this transition in two steps. And in my opinion SW made one huge mistake, and that's the repricing structuring based on a (silly) cheapest print orientation setting. Nobody asked for a default wrong print orientation. My guess: at least 75% of the products on sale at SW is now positioned in an unfavourable print orientation. And why is that?
    PR-wise I think it was not clever to communicate a price drop for 70% of the products based on a quite different print orientation.

    Please look at this topic:
    where I learned on May 10th that the price reduction was actually based on a different print orientation. Now I mentioned that that was not a wise thing to do back then and I still stand behind all my comments so far.

    If SW would go for a better print quality and less needs for reprints (what I assume is the rationale behind this change), then why starting with low price promises based on faulty conditions and then later implement the wanted tool - but - with the wrong condition as starting point.

    Currently SW is not providing very usefull tools for shops with a big inventory and basically sets back the clock.

    Last edited: Jul 17, 2017
  11. ETS35
    ETS35 Well-Known Member
    @railNscale now I get it. It is the amount of work that is the problem, not the actual options offered in the tool. That's a fair argument.
    Yet I do not agree on test prints becoming worthless. They had their value in the past, but now they are not really needed any more for orientation. They will still tell you how good the design is. I will continue to order test prints to battle rejections for my customers. Test prints remain important.
    (BTW I've seen the same file be printed in up to three different directions, so pre-ordering never guaranteed directionality on subsequent prints anyway)

    As I've stated before I would like Shapeways to allow designers to set their standard orientation to 'as is in the file uploaded', rather than the cheapest option. This can also be applied to existing files as well. Problem solved. Are you listening Shapeways?

    BTW, if you quote me I'd appreciate it if you quote me in full. ;-)
    I stand by this, even though it takes some learning. BTW, complaining about poor prints has always(!) worked for me. In that sense you could control it, but only after getting the printed items first... not good for Shapeways or the designer (and even worse for the customer). This should now be a thing of the past. Better late than never.

    I do fully agree that Shapeways repricing information was misleading by calculating the cheapest printing option. I am happy the additional damage this did to the prices of my designs was limited. I am lucky to only have a limited amount of designs in my shop and most of those already matched the cheapest option. Checking all the designs to make sure and correct them was not fun though!
  12. Anbar
    Anbar New Member
    One question.

    When a new part is uploaded the Sw system assigns it a cost in FUD, let's say £10.39p and set the model to SWs preferred orientation.

    When I go into the "Set Orientation" view mode though the cheapest option is £13.50p

    If SW is choosing the "Cheapest Orientation" for the default price, why is the price different between the default (initial) price given by the system, and the "Cheapest price" given within the Set Orientation window?
  13. ETS35
    ETS35 Well-Known Member
    @Andrewsimonthomas, at least one of my designs refuses to save with the right orientation. Is this a known bug?
    Last edited: Jul 18, 2017
  14. PenistoneRailwayWorks
    PenistoneRailwayWorks Well-Known Member
    I believe the tool shows the price with VAT added whereas the edit page doesn't. It would be much better though if the two views were consistent,

  15. MitchellJetten
    MitchellJetten Shapeways Employee CS Team
    @ETS35 can you share the SPIN number / Model ID?
  16. ETS35
    ETS35 Well-Known Member
    That would be 6290887
    I'm concerned about this bug, because now I may have to check all the designs I though I fixed. It would be helpful to see easily see which designs I have already given an fixed orientation.
  17. MitchellJetten
    MitchellJetten Shapeways Employee CS Team
    Thanks @ETS35 :)
    It's a known issue which will be fixed as soon as possible
  18. MitchellJetten
    MitchellJetten Shapeways Employee CS Team