GemologyOnline.com
https://www.gemologyonline.com/Forum/phpBB2/

Another "new gem design app" thread
https://www.gemologyonline.com/Forum/phpBB2/viewtopic.php?f=8&t=23248
Page 12 of 24

Author:  Richard Burton [ Tue Sep 05, 2017 10:05 pm ]
Post subject:  Re: Another "new gem design app" thread

In 'tilt performance' I would like to be able to adjust 'frames per min'.
Ability to resize the performance window.
Have some 'performance number' that preferably I could set the parameters for. Having said that. I still have a hard time understanding the gemray system of allocating importance to cos, table etc.
Any possibility of including dichroism,
Deliberate positioning of inclusions, either silk clouds. veils or cracks.
I also really like the bog ability that shows a map of different pavilion angle performance

Plus I would like to understand more about your clarity function, color density versus saturation.
I think it's a great program that could easily become the defacto raytrace for gems

Author:  Jclimactic [ Tue Oct 03, 2017 5:51 pm ]
Post subject:  Re: Another "new gem design app" thread

Rej wrote:
...

Again, thanks for the testing, I'm glad to finally have it in people's hands. Now begins the fun part of fixing things, putting out new builds and narrowing in on a release!

Has anyone incorporated the app into their workflow for anything yet (TanRatio maybe?). I know there are still issues for building new cuts from scratch, just curious... :D


Rej: Just want to say thank you so much for this program! Just to give you some feedback on that question (from a while ago): the stones in this bracelet were designed in Gem Cut Studio, including using the Tan Ratio function and optimizing angles based on an eyeball review of the in-program rendering. The 0.9v beta worked great as is, for this relatively simple cut.

Not trying to be pushy, but I'm really looking forward to an updated version :D

Attachments:
31202240_Unknown.jpg
31202240_Unknown.jpg [ 48.32 KiB | Viewed 1142 times ]

Author:  Rej [ Tue Oct 03, 2017 9:29 pm ]
Post subject:  Re: Another "new gem design app" thread

Wow, beautiful work. =D> It makes me happy to know my app had a part in it.

The last few weeks were school holidays here in Aus so we got away from screens to hunt rocks again but I'm back at it (right now, actually)

I'm REALLY trying to push to have v1.0 done by xmas, and want to get another test version out before then.

Author:  justinkprim [ Tue Oct 03, 2017 11:37 pm ]
Post subject:  Re: Another "new gem design app" thread

best xmas gift I can think of!

Author:  PandaP [ Mon Oct 23, 2017 8:17 pm ]
Post subject:  Re: Another "new gem design app" thread

I think I found a bug! You might know of it already, but it seems like a gem design with more than 35 tiers of facets won't let you scroll down farther. I'm adding in cutting instructions for the "Super Pear 96" and it will display the first 35 facets but the scroll bar runs out of room to display the last few crown facets as well as the table.

Thought I'd add it to the pile!

Author:  Precision Gem [ Wed Oct 25, 2017 1:55 am ]
Post subject:  Re: Another "new gem design app" thread

PandaP wrote:
I think I found a bug! You might know of it already, but it seems like a gem design with more than 35 tiers of facets won't let you scroll down farther. I'm adding in cutting instructions for the "Super Pear 96" and it will display the first 35 facets but the scroll bar runs out of room to display the last few crown facets as well as the table.

Thought I'd add it to the pile!

Glutton for punishment. Maybe 10 total tiers is the most I would want to get involved with.

Author:  Shifter55 [ Wed Oct 25, 2017 4:08 am ]
Post subject:  Re: Another "new gem design app" thread

PandaP wrote:
I think I found a bug! You might know of it already, but it seems like a gem design with more than 35 tiers of facets won't let you scroll down farther. I'm adding in cutting instructions for the "Super Pear 96" and it will display the first 35 facets but the scroll bar runs out of room to display the last few crown facets as well as the table.

Thought I'd add it to the pile!

I reproduced this on my own machine PandaP, and the bug does a bit more than just overlapping very long faceting instructions. It will cause the program to stop responding if you use the workaround in the explanation below to select the bottom-most tier, among other things like the Tier Edit Menu not overriding the cutting instruction table properly sometimes.
That's a pretty interesting bug you found!

It seems to be due to the Tier Edit Menu acting like a dropdown menu that is locked to the bottom of the Instructions sub-window, rather than to the cell of the table that the cutting instructions are displayed in.
So when a tier is selected in the sub-window it generates the dropdown menu and overrides the parts of the cutting instructions table it overlaps and ignores the scrollbar like a regular dropdown menu does to the menus it overlaps.
If you unselect the tier you clicked on, the dropdown menu should go away and you can access the area that the Tier Edit Menu was overlapping, but the since the only way to unselect a tier in this is to delete the selected tier, you have to create a new tier and then delete it to access those overlapped tiers.

Author:  PandaP [ Wed Oct 25, 2017 12:24 pm ]
Post subject:  Re: Another "new gem design app" thread

Precision Gem wrote:
Glutton for punishment. Maybe 10 total tiers is the most I would want to get involved with.


Haha maybe, I'm only a year and a half into cutting so I feel like doing more complicated stuff is how I'll get better. Since it's a pear it's a lot of tiers of two facets.

Shifter55 wrote:
I reproduced this on my own machine PandaP, and the bug does a bit more than just overlapping very long faceting instructions. It will cause the program to stop responding if you use the workaround in the explanation below to select the bottom-most tier, among other things like the Tier Edit Menu not overriding the cutting instruction table properly sometimes.
That's a pretty interesting bug you found!

It seems to be due to the Tier Edit Menu acting like a dropdown menu that is locked to the bottom of the Instructions sub-window, rather than to the cell of the table that the cutting instructions are displayed in.
So when a tier is selected in the sub-window it generates the dropdown menu and overrides the parts of the cutting instructions table it overlaps and ignores the scrollbar like a regular dropdown menu does to the menus it overlaps.
If you unselect the tier you clicked on, the dropdown menu should go away and you can access the area that the Tier Edit Menu was overlapping, but the since the only way to unselect a tier in this is to delete the selected tier, you have to create a new tier and then delete it to access those overlapped tiers.


Thanks for double checking my bug checking! I tried to re-organize the tiers through the "move up/move down" command but it got too complicated. Plus I'm almost done this stone so I figured I could bite the bullet and go off the PDF. Amazing how fast we become accustomed to new tools and programs haha.

Author:  justinkprim [ Thu Oct 26, 2017 3:35 am ]
Post subject:  Re: Another "new gem design app" thread

I've been playing with the app a lot today to teach myself more about optics. Now that I've built up a number of stone designs, I'm trying to optimize the designs for maximum light return. There are two great tools: Scale and Tilt Performance.

Feature request (maybe this has already been mentioned): It seems like once you finish a design that you like you should be able to click a button the optimizes the design for maximum light return for the gem material. By using Scale, you can optimize by eye but it's pretty hard to tell everything based on the render. Since the app knows light return using the Tilt Performance tool, is there an easy way to adjust the Scale automatically while measuring Tilt Performance? I want the app to find the best ratio's for me because that seems like it would be faster and more accurate than my eye.

Tell me if thats easy or hard ;)

Still shooting for an xmas release? This would be a great gift for my girlfriend.

Author:  Rej [ Thu Oct 26, 2017 5:47 am ]
Post subject:  Re: Another "new gem design app" thread

You're right that I have all the info required to add a performance optimizer already, two things are missing:

1 - Correct metrics of what is "good". Is brightness that important, especially if it's at the detriment of other things like contrast or scintillation? A stone that acts as a pure white reflector would score very well on such a simple test yet would produce a boring looking gem... There are other metrics that need to plug in, and a way to set your priorities. Gemray gets it pretty close but I've heard many complaints that its metrics (at their defaults) always produce shallow crowns because that's what gives the highest score. But yeah, it's just a matter of figuring out which things to measure, giving the user a way to tweak those and then let the processor run an large combination of tangent ratio possibilities and measure the result.

Another approach is a "semi-automatic" way where the computer runs through all the permutations for a range of angles, and shows you the result (in a grid), a bit like the "Merit function mapper" in BOG, but maybe with actual raytraced results so you can judge yourself which one looks "good". This might be a first step, before a full automatic optimizer is added.

2 - Time. All these things take time to code (obviously), even if I have a good idea of how to do it. But yes I'm still focusing on xmas release and a beta hopefully next week.

Author:  PandaP [ Thu Oct 26, 2017 12:07 pm ]
Post subject:  Re: Another "new gem design app" thread

Rej wrote:
Another approach is a "semi-automatic" way where the computer runs through all the permutations for a range of angles, and shows you the result (in a grid), a bit like the "Merit function mapper" in BOG, but maybe with actual raytraced results so you can judge yourself which one looks "good". This might be a first step, before a full automatic optimizer is added.


Awesome info Rej, I think this option would be fantastic. Even if there was an "optimizer" that made it exactly to my set metrics, seeing a range displayed seems like a really great way to at least feel like I have the illusion of choice ;)

Author:  Faceting 101 by Arya [ Wed Nov 01, 2017 4:30 pm ]
Post subject:  Re: Another "new gem design app" thread

Rej wrote:
Another approach is a "semi-automatic" way where the computer runs through all the permutations for a range of angles, and shows you the result (in a grid), a bit like the "Merit function mapper" in BOG, but maybe with actual raytraced results so you can judge yourself which one looks "good". This might be a first step, before a full automatic optimizer is added.

Rej, I think having this map/grid thing is probably the best way to do it. Designers are a lot more likely to spot "special effect" patterns having one of these giant displays, as opposed to using some kind of built-in optimization tool. Plus, let's say you have an optimization tool that allows you to set weights for various parameters, like GemRay does. Who's to say that the designer wouldn't have found a variant better than whatever local/absolute maximum the optimizer finds?

It's pretty easy to look at a whole bunch of crown and pavilion scaled renders and pick one, than trying to figure out optimization weights and all that.

Author:  justinkprim [ Thu Nov 02, 2017 11:18 am ]
Post subject:  Re: Another "new gem design app" thread

YES! Semi Automatic please :smt007 :smt007 :smt007

Author:  Shifter55 [ Thu Nov 02, 2017 10:14 pm ]
Post subject:  Re: Another "new gem design app" thread

Faceting 101 by Arya wrote:
Rej, I think having this map/grid thing is probably the best way to do it. Designers are a lot more likely to spot "special effect" patterns having one of these giant displays, as opposed to using some kind of built-in optimization tool. Plus, let's say you have an optimization tool that allows you to set weights for various parameters, like GemRay does. Who's to say that the designer wouldn't have found a variant better than whatever local/absolute maximum the optimizer finds?

It's pretty easy to look at a whole bunch of crown and pavilion scaled renders and pick one, than trying to figure out optimization weights and all that.

In addition to being easier to work with a grid of static images, a MFmap is also necessary for optimising low-RI materials like opal and fluorite. Trying to balance overall brightness with table brightness without user supervision tends to end badly due to the optimal brightness peaks in those materials sometimes being in areas that create windowed designs.

Plus, it lets you see how much room you have in the crown before the edges start to window, handy when you want extra scintillation around the edges or want to avoid dark corners.

Author:  Rej [ Mon Nov 06, 2017 7:17 am ]
Post subject:  Re: Another "new gem design app" thread

Well, you'll be getting your beta a few days later than I would have liked but there seems to be lots of interest for this feature...

I'd been on the fence about including some kind of optimizer in the first release so this puts that issue to rest: a "manual" tangent ratio optimizer using face-up renders.

Image

Image

Image

Watcha think?

Page 12 of 24 All times are UTC - 4 hours
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/