Update - Planned Features |
(07-05-2021, 02:34 PM)MichaelCollins Wrote:(04-23-2021, 10:37 AM)Ted Woods Wrote:(04-23-2021, 01:44 AM)JLaird Wrote: I did a quick search but did not find any information on what features the next update will include. Also is there an area where you can vote on future suggested features? Hi Michael Thanks for your interest in Kubla Cubed. Like you say in Kubla Cubed 2019 it was not possible to extract vectors from PDF. However this changes in Kubla Cubed 2021 : In Kubla Cubed 2021 we don't do the 'click\stroke to select' auto-tracing as seen in some other products. Instead we try to select all the contours by their shared characteristics (PDF layer, colour, thickness etc...) In theory we believe this is a superior system, in effect trying to use advanced filtering to bulk select the contours instead of manually stroking or clicking on each one. If you have 200 contours it will be one operation instead of 200. Ideally you will then end up with a lot of contours at '0' elevation. This is usually solved in the individual clicking method by prompting for each Z value in turn, however we think we have a better solution with our 'Set Multiple Elevations' tool, where you can set groups of contours all at once. That being said Kubla Cubed 2021 V6.0 will feature some pretty new technology and I wouldn't expect it to be a match for the far more mature solutions other software offers. I think over time, as we iron out the kinks, we hope it will be a superior workflow as our CAD workflow and PDF workflow will use the same tools. It's an interesting subject. In theory you could use all sorts of tricks to extract contours some of them on the cutting edge of computer science, such as : Optical Character Recognition (OCR), Edge Detection Algorithms (to convert a raster to Vector) even machine learning. However I don't know how far down the road other software is with this. We are certainly at the beginning only extracting the vectors. The other solutions are very complicated and will probably be quite hit and miss. Most of them work 'some of the time' which is often no use to an estimator if for instance have to go back and double check each contour level or point. In regard to pipelines the new setup is much improved. We spent quite a lot of time designing the new trench, berm, path behaviour. This can do stuff like trench layouts, backfilling etc... very nicely. We don't incorporate actual pipes though. Kubla Cubed has no concept for utilities or structures. You could create a berm in a phase to fill the area of a pipe and then fill over the top of that in a subsequent phase to calculate the backfill. A lot of people do this also with structural elements. Create in affect dummy earthworks and then fill around and against it in a subsequent phase. I would say our trenching (which comes included in Kubla Cubed pro, not an extra module) is superior to many in the market. Feel free to request a free Trial of KC2021 when it is released.
That all sounds amazing Ted! I agree, the one operation over the 200 is going to be a time saver, even if we still have to input the Z value. Unfortunately, like any program, and the files we try to open within them is the quality of the initial PDF. We have had dashed contour lines that are connected vectors, so each dash is treated as a separate contour unless connected somehow within the program.
I've signed up for the webinar for the 2021 launch and will definitely download a trial once it becomes available. (07-05-2021, 05:10 PM)MichaelCollins Wrote: That all sounds amazing Ted! I agree, the one operation over the 200 is going to be a time saver, even if we still have to input the Z value. Unfortunately, like any program, and the files we try to open within them is the quality of the initial PDF. We have had dashed contour lines that are connected vectors, so each dash is treated as a separate contour unless connected somehow within the program. Hi Michael We have released Kubla Cubed 2021 V6.0 complete with PDF vector extraction. It is a V1 technology so plenty of room for improvements. Would be interested in getting your feedback from it. We have support for Dash lines and matching PDF text to points. Plenty of improvements left to do in subsequent iterations but the fundamentals are there for us to build on. |
Users browsing this thread: |
4 Guest(s) |