DIGITIZE YOUR BEST MODELS UP TO 10’4"
RECEIVE A SHAPE 3D FILE THAT YOU CAN CUT ON ANY MACHINE THAT RUNS ON G-CODE (KKL, APS3000, DNS, 3DM)
YOU WILL OWN THE FILE NOT THE CUTTING HOUSE
RESIZE YOU FILE, MODIFY OR USE AS A BASE TO CREATE MORE MODELS
ONE WEEK TURN AROUND
CONTACT: info@asisurf.com for scheduling and pricing
WE DO NOT SCAN UNLESS PERMITTED BY ACTUAL SHAPER
SHAPE 3D SOFTWARE VERISON 8 AVAILABLE FOR 500 EURO (DESIGN) FROM ASI AND RECEIVE TECHNICAL SUPPORT
surfding, what method of digital scanning are you using? are taking actual 3D data directly from the surface? or manual measuring and transfering the data manually into shape 3D?
I actually use a digitizer and register the information into a csv file then translate it into Shape 3D. My scanner works with servo drives. Manual scanning is not as acurate. We can scan for any cutting house as well as board companies… I took nearly a year to develope, however it was well worth the effort.
Some machine operators are actually using the scan terminology for manual plot point method of connect the dots. Or using cross sections to create surfaces and calling this as scanning. This is not considered a scan. But a created surface that has deviated from the actual board being digitized.
Are you actually writing the tool paths over the board then scanning in slices in cross sections then transferring the data to 3D to create surfaces in shape 3D?
Or, taking plot points then creating surfaces from the plotted data from the digitizer into shape 3D and create surfaces from the plot points?
Either of the methods above are not true scans even if a digitizer is utilized. I have experimented with all of those methods and both are not accurate to the board being digitized. The milled board will be more of a facsimile the discrepancies will be evident on the rail contours and outline. From my experiments the only true method of accurate scan is to write tool paths over the entire board and pick up the complete surface in 3D with out creating surfaces. In this method there aren’t any weird issues with algorithms and scaling, it will scale infinitely with out any distortions. Just my observations.
You would be suprised how well our Digitizer works with the new Shape 3D scan conversion software.
If you know how to use Shape 3D correctly you should be able to generate tool paths without distortions.
Before you cut a file you can clean it up and make sure their are no glitches. I had some issues in the beginning.
It did take some time to work out some of the bugs. Now we can reproduce a model in a couple of hour.
DIGITIZE YOUR BEST MODELS UP TO 10’4"
RECEIVE A SHAPE 3D FILE THAT YOU CAN CUT ON ANY MACHINE THAT RUNS ON G-CODE (KKL, APS3000, DNS, 3DM)
YOU WILL OWN THE FILE NOT THE CUTTING HOUSE
RESIZE YOU FILE, MODIFY OR USE AS A BASE TO CREATE MORE MODELS
ONE WEEK TURN AROUND
CONTACT: info@asisurf.com for scheduling and pricing
WE DO NOT SCAN UNLESS PERMITTED BY ACTUAL SHAPER
SHAPE 3D SOFTWARE VERISON 8 AVAILABLE FOR 500 EURO (DESIGN) FROM ASI AND RECEIVE TECHNICAL SUPPORT
The Shape 3D and Aps programs are great tools for visualization, if needed. They are not true 3D programs but rely on algorithms to create missing data from 2D slices. Remember when you were a kid and learn to draw by sketching using connect the dots? How accurate can a scan be by only picking up a few points?
This is the reason why, if you have a keen eye for shapes most all boards that come off machines that use this method look exactly the same. Due to the algorithms, it locks your shapes into where the slices are located. I have noticed this from mainly Aussie boards, same curve from different shapers. This is now becoming noticeable with the other shaper using these programs and machines here in the states. I can now look at a board and tell what program the board was designed and milled from. In the past this would never happen when shapers design and mowed the “ol’ fashioned” way. You would never see two exact curves and rail contours from two different shapers even if the boards are the same general dimensions.
Any type of method that utilizes cross sections to connect the dots is relying on algorithms. Depending on the algorithms some are better than others but in general create a nice smooth albeit created surfaces and tool paths. In contrast the slices method aren’t accurate to a true 3D scan of the entire board/surface, it’s relying on the slices to full in the missing surface.
The mathematical formulas behind those 2D programs are designed to “average out” the missing data between the slices when creating a board. (not scanning) All created surface programs such as Shape3D, APS, DSD, SurfCAM etc… rely on algorithms.
If using a digitizer to pick up points to create slices with a standard stylus probe with out tool compensation you are not able to recreate the correct shape/contours, esp if you are using a custom cutting tools, even though a digitizer is used it’s still an averaged out surface due to relying on slices. This method is not a true 3D scan. The data is much more accurate than using the manual measure and input of your cross sections method. But still the digitized data is still relying on algorithms to create the missing data between the slices.
The only method I have found to do a true and accurate 3D scan is to not rely on algorithms but to digitize the entire surface of the board. Doing it this method ensures that the shapes will be original to the shaper and not the program.
Do you have this type of scanner? What does it cost?
Surfding, yes I do have this type of scanner. I can’t remember the exact cost, but it was not cheap. If you already have a good digitizer, any decent machine that has a CNC controller with programmable capabilities should be able to do FULL 3D scans. My personal experience I had to learn the program language to be able to edit and write post processors to get excellent results. I’m sure who ever built and designed your machine would assist you in this area. As I am not sure what type of machine you own and the capablities. From what I have seen most over the counter machines are not very flexible in this area. You get what you get and that’s pretty much it. To do any type of modifications like this would be somewhat costly endeavor.
Also, I had the need to have some basic knowledge of how to use a good CAD, CAM programs and edit them. For myself, having absolutely no knowledge of this aspect, it was a long learning curve.
I have used and experimented with most all types of digitizing methods to recreate shapes and there is absolutely no other method that comes remotely close to a full 3D scan.
The partial scan of cross section and created surfaces method is probably the closest to a full 3D scan but its still relying on created surfaces to fill in the missing data between the cross sections. This method of digitizing is great for a beginner or less experienced shapers that may need to clean up a lumpy/bumpy shape that may appear in shape that is being digitized. This method of digitizing will create a smoother surface than the actual board being digitized due to the averaging out of the data. Due to the limited data collected and relying on the algorithms to create a smooth surface across a few slices or cross sections. To put it bluntly, a nice way to clean up a crappy shape. Many less experienced shaper have become overnight masters.
Strange thing between a beginner and old master shaper is both have lumps and bumps in their shapes, it’s just the old master has them placed in the overall design by intention or maybe unintentional but for some reason the board is magic. By design or not, there are many subtle deviations in the magic shapes that are massaged by hand into the foam to where the slice method is not accurate and doesn’t pick up enough sufficient points to recreate the true surface. I have analyzed and scanned hundreds if not thousands of master/magic shapes for many different shapers and have used different scanning methods and the only way to recreate the magic is with a full 3D scan. Not all scans are the same! Full 3D scan is the only way to recreate the magic in shapes with exacting detail without relying on algorithms to create surfaces.
Curioius, Have you tried any other methods?
I have been scanning a lot of boards the last few month with a very good success rate. The thing I like most is our ability to scale the model up or down without distortion. With tracing scanners and the software they work with there are distortions when you try to scale them. Shape 3D is great for scaling. I have a few Pro’s that I work with were we have repeated their magic carpets over and over. When we get their perfect model scanned and converted into a program we can design a series of boards off that same model.
Currently I am working with an Scanning Engineer who has built $500,000 laser scanners for Defense Contractors. The Laser scanner we are delevoping will scan a surfboard in less that 15 minutes. Our next step is to translate the scan into shape 3D. Last night I took a generated file and read it in Rhino 3D. I was then able to open it in Shape 3D. We will simplify this process were we can translate directly to shape 3D. Once this is worked out I will have a Laser scanner off line from my CNC Machine so I can cut and scan at the same time. In the meantime my Digital Scanner is providing very good results.
I am curious with your results with the laser scanner. The laser method was my second method of experiments with digitizing a board. It was less than stellar results. I spent time with the tech from the laser company spending many hours and we could not get a consistent and satisfactory result. There were some optical and tracing issues that couldn’t be resolved, I have a feeling it had something to do with the laser reading into/onto the glassed surface picking up incorrect data and sending the tracer into quagmire. Also, I believe the general shape of a surfboard could have been a problem for the laser esp when it would have to read all the way to the edge of a rolled surface such as the edge of the rails or subtle contours changes. I eventually went to back to the touch scanner type and learned to write the program for FULL 3D touch scanning.
Man, what a long and winding road. Although the journey has been bumpy it’s been quite a ride.
The Shape 3D and Aps programs are great tools for visualization, if needed. They are not true 3D programs but rely on algorithms to create missing data from 2D slices. Remember when you were a kid and learn to draw by sketching using connect the dots? How accurate can a scan be by only picking up a few points?
Pretty darn accurate? You are not playing connect the dots here, you are approximating curves with curves algorithms where you have an extreme level of control of the curve using only four control point for each curve. There are algorithms that will ensure that the curve has the least error possible for every sample point. Given fewer sample point, a bezier curve is going to be a closer approximation of the curves of any given surfboard compared to a ‘low’ resolution mesh of sample points which is indeed a much closer analogy to connect the dots compared to using a higher order function for approximation, and have issues when it comes to making even minor design changes that require above average understanding of computer graphics, methods and tools. On the other hand, as frequency sample rate increases a mesh will approach and surpass approximation with longer curves if there are details that the longer curves will not
In theory after a scan with say one sample at every half inch in every direction, a little higher on the rails you could generate a shape3D file and check the accuracy for every single sample point. Then it’s a question of what accuracy you need, how you interpolate, using adaptive methods to modify the approximation algorithms to get the results as close as possible(ie. adding shorter curves and cross sections where the worst errors are), if it needs a human touch to get the correct accuracy, if you want or need an asymmetrical board, etc. Given that the difference between the scan and the file is approximately the same as the machine accuracy I’d say it is ‘good enough’. The only way to prove it is to cut it though.
I reality I think that design software like shape3D, aps/aku, etc. is a completely different way of designing a surfboard, in reality you start with a computerized design instead of making a copy of a plug. However, if you do have a plug and want that as a starting point you may want some help to get that starting point as accurate as possible.
regards,
Håvard
haarvard, sounds good in theory but that is not the results. how many boards have you scanned? and how many have you milled and compared?
Thanks for Jumping in Haavard.
My Engineer friends who design Airplane wings with digital frame scanners (Touch Probes) like Shape 3D for the same reasons you mentioned. I have scanned 100’s of boards with great success and close to 4,000 cuts with this method. Most of the pro’s on tour in fact ride board’s made from these medthods. Shane Dorian’s boards are all from Shape 3D and cut on a 3DM. There’s the AKU Shaper - APS300 (How many Pro’s boards in oz), Even the DNS (Brazillian Machine) Eric Arakawa has designed some of the best boards in world for many top Hawaiians and don’t forget Pat Rawson, Cino, DOC, Biolas, Patterson to name a few who all use Design Software. There are many methods and I think we should all keep an open mind and respect each others approach to board design or reproduction. There are people who have a hard time using design software and that sometimes can be a stumbling block for them. You have to be somewhat of an artist to control these programs just like with actually shaping.
Surfding, most engineers don’t understand the subtle nuances in surfboard shapes will guide you down this cross section slice method. They don’t understand the difference between an airplane wings and a fine tuned surfboard that has extreme subtle tweaks that aren’t even noticeable to the eye. To them it’s all the same. NOT.
This was my first attempt and methods to digitizing. Like you I had inquired about different methods of scanning and digitizing surfaces from aerospace engineers and computer modelers. Every single one sent me down this road except for one that sent me down the laser scan method. (that was another ordeal) As you know, said method will cut you a board, but not to the “resolution” of my masters and study models being digitized. Good enough? Well maybe for some. This is when I realized that I needed to look into different methods.
I have many unsatisfied customers that have had their master so called scanned from the machines that you have mentioned above and it is clear it is not remotely close to the so called scanned board. The shapers that you mentioned are great foamsmiths machine or not, as most own these machines. Most are too busy shaping and have absolutely no knowledge of what is out there. Maybe second hand knowledge, but most just make work with what they have.
I am sure the shapers in the know would be the first to admit that the machines they own don’t have the capabilities to scan at the same resolution as a full 3d scan. Once they own it, most are just happy to get boards cut.
For kicks, (and maybe some wager ;)) I would love to do a comparison of scanning techniques.
Lets’ say I have a 5’10” x 21 x 2.625” (using bottom length and width measurements)_super fish magic something or other, and joe ripper wants the exact board but wants all his exact foot dents shaped in the exact location as the board being scanned, along with that the shape has a subtle high spot (dish nose and tail, step, hump whatever.) beak nose, deep swallow, with triple wingers, forward roll, to subtle single to double out the back.
Now, Joe rippers dad wants the exact same board from the same scanned board but would like to change the dimensions slightly, he want to go 7’10” x 23” x 3” thick but no foot dents on the deck.
We both scan using our methods, both scanned in a normal work day 8-10 hr day.
Then we cut both boards and compare.
and ill join in the competition with a digital micrometer and straight edge and design software… while your still scanning i will have the shape copied and finished in 2-3 hours. with a guarantee to work as good or better (note work/surf better, not measure the same to the umpteenth degree) than the origional. (one catch , i want complete control over the foam, glass etc i use to finish the board)
lol…you must have one fast straight edge. you on! when?
Hi Oak,
No, I have not done any scanning. I’m a nobody when it comes to ‘the subtle nuances in surfboard shapes’. However, what I wonder is why said theories does not work. Is it only based on experience? Because in that case there might be a lot of things that can go wrong in the process and the programming. One little detail and everything falls apart. And there are certainly many little details that can and you’ll probably have Murphy with his laws on visit a lot. What I would like to know is what problems you’ve run into using said methods and why it will fail.
The case you mentioned is obviously the hardest for a method that rely on interpolation as interpolation wants to make a perfectly smooth board. I’m still puzzled though why you’d want a board that is not smooth. I keep hearing of magic boards with imperfections, but I have yet to hear any rational explanation of why this (and what) works better. If there was a logic behind it, these imperfections could be utilized and incorporated into the designs. Also, unless the guy who is going to finish the board is extremely delicate with his sanding and the sander after the glassing is equally delicate, a lot of these micro nuances are gone before you before you have a finished board.
In the end, what we’re up against is a board shaped by hands and a planer. Making a perfect copy of a board full of dents and imperfections is not going to happen by hand either, if that’s really what you want.
regards,
Håvard
Håvard, from my experience this is where the mystical “magic” is. If you come from a handshaping back ground with some mileage then you would understand what the difference is. It’s the process of the elusive replication of your special “magic” boards. Smooth is fine, but it’s not necessarily what your after as a hand shaper it’s more about flow and feeling not an overly algorithmed mathematically perfect shape that is extremely smoothed out. That’s not what I’m after.
I know it’s hard to understand foot dents in new boards in this age of pop outs and ridgid constructions. Yet these are some quirky requests that I have had to deal with. Some advanced riders like the boards to be an extension of their body, the foot dents allow the deck rocker and deck surface to be below the surface of the feet. The foot dents allow the rider to have more control over the board in extremely critical situations, much better control of the board than just a smooth deck surface. This is like ordering a custom set of shoes custom made to your own feet, kinda like pro athletes having custom made shoes for competition. If you have ever shaped custom boards for guys of this caliber they will all tell you, “I can’t wait to break it in and get my foot dents in it.” Strange but true.
From my experience there is more to a magic board than just smooth. Interpolate = interpret. Interpolate meaning,
1.
to introduce (something additional or extraneous) between other things or parts; interject; interpose; intercalate.
2.
Mathematics. to insert, estimate, or find an intermediate term in (a sequence).
3.
to alter (a text) by the insertion of new matter, esp. deceptively or without authorization.
4.
to insert (new or spurious matter) in this manner.
Interpret from point to point. Sorry, from my expererience something get’s lost in translation. (kinda like using google to translate japanese 
Soemthing lost get in translation. hehe…How can you call creating surfaces and call it a SCANNING to a degree of any accuracy with only 3 slices? I think people need to know the algorithms (interpolations) are pre set mathematical formula, ie shaper A and B would get the same results by punching in the same data in the same 3 slices. This is the reason why all the board that come off these machines all look alike because they are from the same formula. Am I the only that can see the difference? Or is everyone that uses these machines and program blind to see the similarity in all the shapes?
Hi,
unless you are able to sample said board into infinite tolerance you have to do some kind of interpolation too, you need to fill in the blanks between your sample points. What kind of interpolation are you using?
The surface algorithms used in the various software out the are actually quite flexible. When I say smooth, it’s all about flow, not necessarily the curve from one tangent to another with the least curvature. A bezier passing though two points can have almost infinite variations and the level of control given the easy of use is excellent. Some would even consider the bezier a somewhat imperfect algorithm mathematically because it’s hard to make it mathematically smooth when you connect several pieces of beziers together, there are splines that do this far better. However, for the level of control and ease of use the bezier is ideal for the purpose of designing surfboards.
Given that you want to reproduce a handmade board accurately, you know as well as anyone that you are not going to get anywhere near it with 3 slices. However give it 12-13 and you are likely to get awfully close. On the other hand, a surfboard made from a mesh of sample points is not a guarantee that it will surf better than one made with 3 slices.
I’m a bit puzzled that you can tell that all board coming from this kind of machines and programs ‘all look the same’, given that there is such a high flexibility in these programs and that the three main programs (SurfCAD, Shape3D and APS3000/AkuShaper) are all using different types of interpolation between the slices, not even going into the design and quality of the cuts made by the machines.
regards,
Håvard
I’m sure OAK is pretty set on his view point. That’s fine. He most likely is a good hand shaper as well? Shape 3D for me is working well and the PRO’s that I shape for love the boards that are produced. In the beginning I struggled with my set up and had my issues to deal with. Yes my scans in the begining were not that desirable. That’s when we re-wrote the scanning software for the Digital Touch Probe. Then Shape 3D wrote the translations software so the scans would work with Shape 3D. Since all this we are on Version 8. Oak, why not try a trial version of Shape 3D V8? I’m sure it will work with your scannner. The thing that I like is the scaling ability of the program. WE are just now completing a LASER scan method that I like because of the time saved. It’s very acurrate as well. My son who is on the tour use to say to me when I would give him a board to ride, “Your no Al Merrick”! That was hard to swallow. But he was right my boards sucked. After working hard making board after board we got closed to what he likes. Finally after $1,000’s of $$$ we nailed it. That’s when I scanned the board that he liked. Cut it shaped it and carefully tuned in his rails. Glassed it then re-scanned it. Now we are able to re-create it over and over. In fact now we are getting request for this same model for many of his friends that he travels around the world with. So I’m having a great time with all this. For me no one can take that away from me.
unless you are able to sample said board into infinite tolerance you have to do some kind of interpolation too, you need to fill in the blanks between your sample points. What kind of interpolation are you using?
The surface algorithms used in the various software out the are actually quite flexible. When I say smooth, it’s all about flow, not necessarily the curve from one tangent to another with the least curvature. A bezier passing though two points can have almost infinite variations and the level of control given the easy of use is excellent. Some would even consider the bezier a somewhat imperfect algorithm mathematically because it’s hard to make it mathematically smooth when you connect several pieces of beziers together, there are splines that do this far better. However, for the level of control and ease of use the bezier is ideal for the purpose of designing surfboards.
Given that you want to reproduce a handmade board accurately, you know as well as anyone that you are not going to get anywhere near it with 3 slices. However give it 12-13 and you are likely to get awfully close. On the other hand, a surfboard made from a mesh of sample points is not a guarantee that it will surf better than one made with 3 slices.
I’m a bit puzzled that you can tell that all board coming from this kind of machines and programs ‘all look the same’, given that there is such a high flexibility in these programs and that the three main programs (SurfCAD, Shape3D and APS3000/AkuShaper) are all using different types of interpolation between the slices, not even going into the design and quality of the cuts made by the machines.
regards,
Håvard