|
|
Post by rmc on Oct 26, 2016 16:29:43 GMT
Hi all, Been a few months since I was here last. I've spent a couple of months working on this program that could help a person calculate the worth of gold based primarily upon the size of the gold in question, rather than having to weigh it. This was thought up by me when I noticed several gold prospectors, seen in various videos over the internet, looking into their pans at the end of the prospecting day wondering how much the gold found is worth. Often, the more experienced miners remark that they can tell its worth, based upon the size of a given grain that they might have seen before. In essence, equating a given size for a certain number of grams (all from experience). An automated version of this may be useful for moments where there are too many tiny grains of gold to try and pick out to weigh, not to mention that a scale may not be all that handy at the moment. Whereas, simply estimating the number of grains and their sizes might be more doable. Also, the likelihood of having internet access out in the field is becoming more and more possible. Just a year ago a friend wrote me from the bottom of Death Valley, and was writing me on Facebook specifically just because he could! So access to this program may be more likely than simply having to have it loaded offline. What I did was looked up the density of gold and the more common impurities that are often found in gold and worked up the math necessary to parse out mass from the various densities. Anyway, I'd like input at this point. (basically, does the program seem intuitively usable? Are the instructions clear? Did I get it correct?) And I cannot think of a more educated, knowledgeable and tactful bunch of guys and girls for getting just this sort of input. The site map at the bottom of the "instructions" page links to "technical dissertation" if you want to see the math laid out completely. Here's the link: hightower.neocities.org/QuickGoldTally |
|
|
|
Post by the light works on Oct 26, 2016 17:00:21 GMT
Hi all, Been a few months since I was here last. I've spent a couple of months working on this program that could help a person calculate the worth of gold based primarily upon the size of the gold in question, rather than having to weigh it. This was thought up by me when I noticed several gold prospectors, seen in various videos over the internet, looking into their pans at the end of the prospecting day wondering how much the gold found is worth. Often, the more experienced miners remark that they can tell its worth, based upon the size of a given grain that they might have seen before. In essence, equating a given size for a certain number of grams (all from experience). An automated version of this may be useful for moments where there are too many tiny grains of gold to try and pick out to weigh, not to mention that a scale may not be all that handy at the moment. Whereas, simply estimating the number of grains and their sizes might be more doable. Also, the likelihood of having internet access out in the field is becoming more and more possible. Just a year ago a friend wrote me from the bottom of Death Valley, and was writing me on Facebook specifically just because he could! So access to this program may be more likely than simply having to have it loaded offline. What I did was looked up the density of gold and the more common impurities that are often found in gold and worked up the math necessary to parse out mass from the various densities. Anyway, I'd like input at this point. (basically, does the program seem intuitively usable? Are the instructions clear? Did I get it correct?) And I cannot think of a more educated, knowledgeable and tactful bunch of guys and girls for getting just this sort of input. The site map at the bottom of the "instructions" page links to "technical dissertation" if you want to see the math laid out completely. Here's the link: hightower.neocities.org/QuickGoldTallynot being a gold miner, I have no knowledge of how easy it is to estimate purity in the field. with that in mind, I am thinking the calculator sounds like part of a solution. as a mass calculator, it is very simple and straightforward. I think adding a regional database for setting the purity setting based on what is usually in the region might be helpful. also, the one gold prospector I have seen in action used a suction bottle for picking up gold out of his sluice - having an estimator that worked on ml volume might be handy for those. |
|
|
|
Post by GTCGreg on Oct 26, 2016 19:08:40 GMT
If all of your assumptions about purity, size, and number of nuggets is correct, then I see no reason why your program couldn't be somewhat accurate.
But here's another idea. Someone, (not me as I wouldn't know where to begin)should write an app that can take a still image jpg or other digital picture file and apply image processing to automatically detect the gold nuggets based on their color and measure their size and then count the number of them. Once that is known, it would be relatively simple to know their value. The imaging program may even be able to calculate the purity based on a color spectral analysis.
|
|
|
|
Post by the light works on Oct 27, 2016 0:01:08 GMT
If all of your assumptions about purity, size, and number of nuggets is correct, then I see no reason why your program couldn't be somewhat accurate. But here's another idea. Someone, (not me as I wouldn't know where to begin)should write an app that can take a still image jpg or other digital picture file and apply image processing to automatically detect the gold nuggets based on their color and measure their size and then count the number of them. Once that is known, it would be relatively simple to know their value. The imaging program may even be able to calculate the purity based on a color spectral analysis. I had thought about that, but the catch that came to my mind was that you then need some kind of scale reference. otherwise you don't know whether its a close-up of small nuggets, or a long shot of big nuggets.. addendum: I also wondered if a color analysis would work. |
|
|
|
Post by GTCGreg on Oct 27, 2016 2:05:38 GMT
If all of your assumptions about purity, size, and number of nuggets is correct, then I see no reason why your program couldn't be somewhat accurate. But here's another idea. Someone, (not me as I wouldn't know where to begin)should write an app that can take a still image jpg or other digital picture file and apply image processing to automatically detect the gold nuggets based on their color and measure their size and then count the number of them. Once that is known, it would be relatively simple to know their value. The imaging program may even be able to calculate the purity based on a color spectral analysis. I had thought about that, but the catch that came to my mind was that you then need some kind of scale reference. otherwise you don't know whether its a close-up of small nuggets, or a long shot of big nuggets.. addendum: I also wondered if a color analysis would work. Right, scaling could be a problem. Maybe you could use some fixed size pan or tray as the size reference and then frame your picture so the entire pan or tray just fills the field. I'm also not sure about the spectral analysis part either. You would also need some kind of color reference or white balance for that to work. Just a brain fart. |
|
|
|
Post by the light works on Oct 27, 2016 3:50:09 GMT
I had thought about that, but the catch that came to my mind was that you then need some kind of scale reference. otherwise you don't know whether its a close-up of small nuggets, or a long shot of big nuggets.. addendum: I also wondered if a color analysis would work. Right, scaling could be a problem. Maybe you could use some fixed size pan or tray as the size reference and then frame your picture so the entire pan or tray just fills the field. I'm also not sure about the spectral analysis part either. You would also need some kind of color reference or white balance for that to work. Just a brain fart. a white card would suit for that - transfer your gold to the card and photograph it - you'd probably want to host the actual software in the site, rather than on the phone. it would then be able to combine GPS tag with color analysis, and use the card to scale the gold. it's getting to be a pretty complicated program, though - a bit higher end programming than the original. |
|
|
|
Post by silverdragon on Oct 27, 2016 8:37:04 GMT
How about a white plastic card with some form of scale on it. Washable, always useful with wet gold?.
Or, include a cube of known size, say a Dice cube to make it multi-purpose, that can be put on any white surface.
Purity is partly seen by colour of the gold, so many miners know from experience the purity.
|
|
|
|
Post by silverdragon on Oct 27, 2016 8:41:22 GMT
Hi all, Been a few months since I was here last. I've spent a couple of months working on this program that could help a person calculate the worth of gold based primarily upon the size of the gold in question, rather than having to weigh it. This was thought up by me when I noticed several gold prospectors, seen in various videos over the internet, looking into their pans at the end of the prospecting day wondering how much the gold found is worth. Often, the more experienced miners remark that they can tell its worth, based upon the size of a given grain that they might have seen before. In essence, equating a given size for a certain number of grams (all from experience). [cut for brevity only] Does what it says on the tin.... Gives you several options to play about with. You done good as far as I can tell. Its easy to work out whats needed. Although I may have some complaints about size of type on a mobile device, ignore me, its just my eyes are older than they used to be.. |
|
|
|
Post by mrfatso on Oct 27, 2016 9:59:22 GMT
How about a white plastic card with some form of scale on it. Washable, always useful with wet gold?. Or, include a cube of known size, say a Dice cube to make it multi-purpose, that can be put on any white surface. Purity is partly seen by colour of the gold, so many miners know from experience the purity. A cube dice would not work, as a RPG and Board gamer I have seen too many of variable sizes to use them as a standard measure. From ones that as so small you can hardly read them to ones the size that you can just about hold in one hand. www.dicegamedepot.com/dice-sizes/The problem would be knowing which size it was to calibrate the app. A reference cube of a known size might be handy, but having been in the filed making on various geological fieldtrips I know how easily such things are at getting lost. Most good compasses and compass clinometers have a scale on the side that are used when reading maps, and those could be used. A credit card size scale that could be put in a wallet is a good idea. |
|
|
|
Post by rmc on Oct 27, 2016 10:03:13 GMT
Silverdragon, interseting that you mention type size as that was a suggestion Taylor (a young woman at school) also made about increasing the size of first-page instructions. As to estimating grain size of the gold photographically and with "face recognition" type software: since I am only just learning javascript, html and css, I'd imagine that sort of next-level version remains out of my reach. And, given that there are people out there who program that sort of thing every day, I'd imagine that if this program becomes the least bit popular, that sort of version would happen pretty swiftly.
Edit:. As far as scale is concerned, I list several commonplace objects for a size reference, perhaps any one of those may help in this regard? Pluck a hair and drop it into view as gold dust scale? Use a matchstick for a 2 mm ref? How does the face rec software judge distance?
Edit II: grain scale of sand could help. It's rough though - no telling if you're looking at silt verses a pebble without the scale nailed down first, I'd guess. Maybe not, though.
See Wentworth Grain-Size Chart.
|
|
|
|
Post by mrfatso on Oct 27, 2016 10:28:49 GMT
Another problem I can think of is that Gold grains are not uniform in their shape, and so when calculating their value if you assume a certain volume for a certain grain size this irregularity will throw off the calculations.
This is something I spent hours studying when attempting to work out the estimated value of oilfields and orebodies for mining, it can take some complicated maths to work out the true volume of a 3 dimensional object compared to what the volume is if you presume it is uniform. The variance that you get between these two figure can mean the difference between profitability and nonprofitability.
I suppose that on the level than you are working on the difference may not make too much difference but there will be a difference in true weight compared to estimated weight.
|
|
|
|
Post by rmc on Oct 27, 2016 11:25:17 GMT
Yep. We went with simple cube. Will be somewhat off for a perfect volume if estimating size by sight. I did make some allowance for entering a volume if you have it.
For distance and size photographically, I'd think once it is in focus, using the focal point, that should give a focal length and, based upon lens type, an angle. Once the software has that, simply trig should work out the lengths of picture frame and whatever is inside it (based on a percent of frame length). So, all you have to know is, is the thing that you zoomed in on in sharp focus?
|
|
|
|
Post by the light works on Oct 27, 2016 13:58:36 GMT
Yep. We went with simple cube. Will be somewhat off for a perfect volume if estimating size by sight. I did make some allowance for entering a volume if you have it. For distance and size photographically, I'd think once it is in focus, using the focal point, that should give a focal length and, based upon lens type, an angle. Once the software has that, simply trig should work out the lengths of picture frame and whatever is inside it (based on a percent of frame length). So, all you have to know is, is the thing that you zoomed in on in sharp focus? problem with that is that I think most phone cameras use pinhole lenses rather than variable focus lenses. I could be wrong bout that. also, considering you are probably going to be doing this with small amounts, it is likely that the distance will be fluctuating by a significant percentage just from camera wobble. I think at this point we are looking at two potential bits of software - one that is designed on your available skillset, and one assuming you bring in a professional to do the coding. I.E. you could provide a searchable reference chart for estimated purities. a professional could build something where they would snap and upload a picture using a calibrated white reference card ($1.99 purchase) and the software host would analyze the picture and return an estimate. |
|
|
|
Post by mrfatso on Oct 27, 2016 14:25:19 GMT
Yep. We went with simple cube. Will be somewhat off for a perfect volume if estimating size by sight. I did make some allowance for entering a volume if you have it. For distance and size photographically, I'd think once it is in focus, using the focal point, that should give a focal length and, based upon lens type, an angle. Once the software has that, simply trig should work out the lengths of picture frame and whatever is inside it (based on a percent of frame length). So, all you have to know is, is the thing that you zoomed in on in sharp focus? Thinking on it it might be worth getting a number of different types of sand, coarse, medium, fine etc. and adjusting the density in the program for sand. Then taking small samples of sand estimate their mass using the program and the weigh them. This would give you an idea of what the variance might be. |
|
|
|
Post by rmc on Oct 27, 2016 15:19:00 GMT
Yep. We went with simple cube. Will be somewhat off for a perfect volume if estimating size by sight. I did make some allowance for entering a volume if you have it. For distance and size photographically, I'd think once it is in focus, using the focal point, that should give a focal length and, based upon lens type, an angle. Once the software has that, simply trig should work out the lengths of picture frame and whatever is inside it (based on a percent of frame length). So, all you have to know is, is the thing that you zoomed in on in sharp focus? Thinking on it it might be worth getting a number of different types of sand, coarse, medium, fine etc. and adjusting the density in the program for sand. Then taking small samples of sand estimate their mass using the program and the weigh them. This would give you an idea of what the variance might be. In a manner of speaking, I may have already done something like this. I was looking for some means to test the software, so I took out my American Gold Eagle (1/10th Oz.) coin, looked up its stats and then attempted to use this information as an input source for the program. Trouble is, it declares that the coin is about 191.2 cubic millimeters or about .19 ml and had an awful time trying to construct a plastic sleeve that the coin could slip into with water, in order to measure the .2 ml change in volume. (about like trying to measure a drop of water in some sort of tank). But, I think I get what you are saying. Do the same program, but with items one is more likely to find commonly and could do more work with. |
|
|
|
Post by the light works on Oct 27, 2016 15:28:09 GMT
Thinking on it it might be worth getting a number of different types of sand, coarse, medium, fine etc. and adjusting the density in the program for sand. Then taking small samples of sand estimate their mass using the program and the weigh them. This would give you an idea of what the variance might be. In a manner of speaking, I may have already done something like this. I was looking for some means to test the software, so I took out my American Gold Eagle (1/10th Oz.) coin, looked up its stats and then attempted to use this information as an input source for the program. Trouble is, it declares that the coin is about 191.2 cubic millimeters or about .19 ml and had an awful time trying to construct a plastic sleeve that the coin could slip into with water, in order to measure the .2 ml change in volume. (about like trying to measure a drop of water in some sort of tank). But, I think I get what you are saying. Do the same program, but with items one is more likely to find commonly and could do more work with. with the sand, I would get a graduated cylinder, and fill it with sand and water, (water level above the sand level and a known mass of sand) then base the tables on the measured volume of the sand in the cylinder. - thus you can estimate mass of wet material. |
|
|
|
Post by rmc on Oct 27, 2016 15:43:04 GMT
linkAlaska gold fineness is discussed in this thread (above link). In all, it may be real work to compile a meaningful list of defined 'fineness' of placer gold deposits from around the world. No particular single work seems to exist currently. But, most agree that fineness generally spans from about 70% to 90%, depending. Also, silver is somewhat more common than copper (both are possible as a blend of impurities) Here's an example of a situation this sort of software could be used. Below is a jpg of someone's pan after gold prospecting. He posted it online. So a person could guess the worth and then guess the worth using some sort of software like this... www.reddit.com/r/Prospecting/comments/59iqly/found_some_gold_finally_at_yankee_jims/?sort=new
|
|
|
|
Post by the light works on Oct 27, 2016 16:45:05 GMT
linkAlaska gold fineness is discussed in this thread (above link). In all, it may be real work to compile a meaningful list of defined 'fineness' of placer gold deposits from around the world. No particular single work seems to exist currently. But, most agree that fineness generally spans from about 70% to 90%, depending. Also, silver is somewhat more common than copper (both are possible as a blend of impurities) Here's an example of a situation this sort of software could be used. Below is a jpg of someone's pan after gold prospecting. He posted it online. So a person could guess the worth and then guess the worth using some sort of software like this... www.reddit.com/r/Prospecting/comments/59iqly/found_some_gold_finally_at_yankee_jims/?sort=new with my "by volume and region" concept he would put it in a measuring tube, then give the volume of gold in the tube. and it could provide him the likely purity from a user accessed table, and calculate the mass based on the volume in the tube. it might even let you bypass calculating the fineness. since it is mixed in fineness. |
|
|
|
Post by rmc on Oct 27, 2016 18:05:46 GMT
linkAlaska gold fineness is discussed in this thread (above link). In all, it may be real work to compile a meaningful list of defined 'fineness' of placer gold deposits from around the world. No particular single work seems to exist currently. But, most agree that fineness generally spans from about 70% to 90%, depending. Also, silver is somewhat more common than copper (both are possible as a blend of impurities) Here's an example of a situation this sort of software could be used. Below is a jpg of someone's pan after gold prospecting. He posted it online. So a person could guess the worth and then guess the worth using some sort of software like this... www.reddit.com/r/Prospecting/comments/59iqly/found_some_gold_finally_at_yankee_jims/?sort=new with my "by volume and region" concept he would put it in a measuring tube, then give the volume of gold in the tube. and it could provide him the likely purity from a user accessed table, and calculate the mass based on the volume in the tube. it might even let you bypass calculating the fineness. since it is mixed in fineness. Depends what's on the "user accessed table". If it provides Karat and impurity, then maybe. Volume isn't enough. Nor is volume and just impurity. |
|
|
|
Post by the light works on Oct 27, 2016 23:44:04 GMT
with my "by volume and region" concept he would put it in a measuring tube, then give the volume of gold in the tube. and it could provide him the likely purity from a user accessed table, and calculate the mass based on the volume in the tube. it might even let you bypass calculating the fineness. since it is mixed in fineness. Depends what's on the "user accessed table". If it provides Karat and impurity, then maybe. Volume isn't enough. Nor is volume and just impurity. my thinking would be that the table provided typical karat and impurity for the region the gold was in. if they know they can enter it directly, and if they don't they can access the table and look it up. |
|
