+janandsteve Posted May 4, 2011 Author Share Posted May 4, 2011 I think Groud Buzzard has the question wrong. He compared multiple units and got the same SS/SR times. That isn't the point of contention here. The problem is that Garmins calculation (I'll assume used on all models) isn't in line with the NOAA, Naval Observatory, and just about every other precision calculator out there. Ummm, no.... He clearly stated that the times matched the NOAA calculation exactly. 'no meaningful difference' = there may be a difference of 1 or 2 mins Quote Link to comment
+janandsteve Posted May 4, 2011 Author Share Posted May 4, 2011 That to me isn't the definition of "exactly". The definition of "meaningful" is subjective. He gave ACTUAL NUMBERS to the second. Man you like to argue for no reason, don't you. yogazoo is merely presenting the FACTS.... the 62s is WRONG, which I agree with, hence why I started the post. 'no reason' - there clearly IS a reason to get this sorted...... the 62s is WRONG. Quote Link to comment
+janandsteve Posted May 4, 2011 Author Share Posted May 4, 2011 If you we're using the calculator function on your unit and input 2+2 and received an answer of 3.95 wouldn't you be concerned? I mean it's close, but who want's to nitpick right? Excellent analogy. To the doubters: Remember if the sunset calculation is out, so is the fish and hunt calculators...... you might as well carry around bits of paper with accurate results, but this sort of defeats the object somewhat..... Just as a sideline...... the Atlantic Bluechart tide tables are also WRONG..... but hey another topic for another day, I guess...... Quote Link to comment
+Ground Buzzard Posted May 4, 2011 Share Posted May 4, 2011 Thanks for confirming that this seems to be an issue across the range of 62 and 400 series. Wonder if the 500 series is indeed the same or has that got the 'correct' formula, considering the test result pasted earlier in the thread. The 450 and 550 run exactly the same firmware. This been the case, I dont understand GroundBuzzard's post and results as below, with ref to the 550t..... ..... I rounded up 9 different Garmin hand-held units.....Oregon 550t (2010)....I ran each one of these at my residence, and noted what the SR/SS times were from each unit for today (3 May 2011). Every last one (NO EXCEPTIONS!) gave sunrise at 06:13 MDT and sunset at 19:52 MDT. These can be compared with the results computed from the standard algorithm (FIRST POINT, above): sunrise at 06:13:14 MDT, sunset at 19:53:18 MDT. I see no meaningful difference from the result given by all the Garmin units. I cannot see it been a location thing - one of the other posters on here tried their 62s in the US ( Montana ) and they get the same discrepancies as what I do. Your tests also confirm the difference.... so I wonder why Mr Buzzard gets his bang on?.....maybe he can enlighten us? My post simply described what I did, and the results I got. I timed the sunset from my buzzardly abode yesterday. The predicted value from the 9 Garmins (see my post) was 19:52 MDT. The observed value was within a few seconds of 19:51 MDT. The apparent difference is qualitatively consistent with the fact that the point of observation, in the Rio Grande valley on the east side of Albuquerque at 5710 feet above sea level, is some hundreds of feet lower than the western horizon, roughly 20 miles away. I do not have the topographic data that would support a quantitative estimate of the time difference to be expected. The matter is further muddied by the atmospheric pressure differing from what the calculation assumes. Perhaps the question should be: How large must the discrepancy be to warrant taking it seriously? Quote Link to comment
+dfx Posted May 4, 2011 Share Posted May 4, 2011 Perhaps the question should be: How large must the discrepancy be to warrant taking it seriously? I got another question for everybody: Do we want to know the times when the sun actually rises and sets, or do we want to know the SR/SS times precisely as they're calculated from some formula? If it's the former, it comes back to GB's question (quoted above). If it's the latter, then the followup question is: which formula? And why this one? Quote Link to comment
yogazoo Posted May 4, 2011 Share Posted May 4, 2011 (edited) Perhaps the question is different for everyone regarding this issue. True, the difference may be small and perhaps negligable for some. For me the question is why doesn't Garmins equation match exactly with the best known mathematical equation to calculate such events? It may not affect practical applications, and I don't synchronize my life with the exact moment the sun sets or rises. It's simply a mathematical/factual annoyance to me and nothing more. Edited May 4, 2011 by yogazoo Quote Link to comment
+JBnW Posted May 4, 2011 Share Posted May 4, 2011 (edited) Actually, going back to Ground Buzzard's last post; maybe he is pointing us in the right direction: It does assume that the observer is at sea level, that his horizon is exactly 90° from the zenith (in other words, that his horizon lies in a precisely horizontal plane), and that standard meteorological conditions prevail. Any deviation from these ideal conditions will introduce a modest error in the result. Such error will increase with increasing distance from the equator, and can become quite significant at latitudes above, say, 50°. Bolding mine. While we are still comparing GPS-predicted SR/SS times versus other predictors (NOAA, Naval Observatory, etc), and not predicted versus reality. Anyway, Janandsteve seem to be the most northerly at N53, no reported altitude, and gets up to 4 minutes per day error. I'll volunteer that I'm at N35, and right at 1000 FASL, and am consistently 2 minutes ahead of the online calculators. I would look to see where Ground Buzzard lives to make an estimate of Lat and/or elevation, but I'd guess he's closer to the equator and/or sea level. And, of course, the site's down now. Anybody else wish to add their's? All that to say, farther north=larger error? PS, and even though I'm in Kansas, our horizon isn't THAT flat, so the real SR/SS would change simply from topography. edit: fix quotey thingy Edited May 4, 2011 by JBnW Quote Link to comment
+dfx Posted May 4, 2011 Share Posted May 4, 2011 While we are still comparing GPS-predicted SR/SS times versus other predictors (NOAA, Naval Observatory, etc), and not predicted versus reality. Anyway, Janandsteve seem to be the most northerly at N53, no reported altitude, and gets up to 4 minutes per day error. I'll volunteer that I'm at N35, and right at 1000 FASL, and am consistently 2 minutes ahead of the online calculators. I would look to see where Ground Buzzard lives to make an estimate of Lat and/or elevation, but I'd guess he's closer to the equator and/or sea level. And, of course, the site's down now. Anybody else wish to add their's? All that to say, farther north=larger error? edit: fix quotey thingy I'm at N 43 20' and at 170m altitude. Oregon tells me SR/SS at 6:09 and 20:22 while http://www.csgnetwork.com/sunriseset.html tells me it's at 6:09 and 20:24. Quote Link to comment
yogazoo Posted May 4, 2011 Share Posted May 4, 2011 (edited) 46.6 degrees North / -112 degrees South (Montana) Elev: 4,000ft GARMIN: SR- 6:10am SS- 8:39pm NOAA (using: CALCULATOR LINK): SR- 6:09am SS- 8:41pm Edited May 4, 2011 by yogazoo Quote Link to comment
+JBnW Posted May 4, 2011 Share Posted May 4, 2011 And some better numbers for me: N39 (mistyped before), GPS elevation 1141 ft GPSMap 76CSx SR 05:25 SS 20:21 NOAA ESRL SR 05:25 SS 20:22 *Note: ESRL reports "Apparent SR and SS Quote Link to comment
+Ground Buzzard Posted May 4, 2011 Share Posted May 4, 2011 Actually, going back to Ground Buzzard's last post; maybe he is pointing us in the right direction: It does assume that the observer is at sea level, that his horizon is exactly 90° from the zenith (in other words, that his horizon lies in a precisely horizontal plane), and that standard meteorological conditions prevail. Any deviation from these ideal conditions will introduce a modest error in the result. Such error will increase with increasing distance from the equator, and can become quite significant at latitudes above, say, 50°. Bolding mine. While we are still comparing GPS-predicted SR/SS times versus other predictors (NOAA, Naval Observatory, etc), and not predicted versus reality. Anyway, Janandsteve seem to be the most northerly at N53, no reported altitude, and gets up to 4 minutes per day error. I'll volunteer that I'm at N35, and right at 1000 FASL, and am consistently 2 minutes ahead of the online calculators. I would look to see where Ground Buzzard lives to make an estimate of Lat and/or elevation, but I'd guess he's closer to the equator and/or sea level. And, of course, the site's down now. Anybody else wish to add their's? All that to say, farther north=larger error? PS, and even though I'm in Kansas, our horizon isn't THAT flat, so the real SR/SS would change simply from topography. edit: fix quotey thingy FWIW, since I find myself near the center of the storm, my coordinates are 35° 08' N, 106° 31' W, elevation 5710 ft MSL. Quote Link to comment
John E Cache Posted May 5, 2011 Share Posted May 5, 2011 My Muslim friends always know when sunset is. Especially important during Ramadan because the daily fast ends at sunset. I was curious how they define sunset. I found this link interesting: http://praytimes.org/calculation/ They account for both height and refraction. Quote Link to comment
+julianh Posted May 8, 2011 Share Posted May 8, 2011 To the doubters: Remember if the sunset calculation is out, so is the fish and hunt calculators...... Which might matter - if there was ANY rational basis in believing the Fish and Hunt Calculator in the first place! I mean - does anyone REALLY think that Alaskan Grizzlies, Indian Tigers, Australian Barramundi, and Scottish Pheasants all follow the same rules in deciding when to come out to feed?! (Or maybe they all carry Garmins in their fur / feathers / fins, and consult the tables to find the best time to forage? ) Quote Link to comment
Now at Zero Gravity Posted May 8, 2011 Share Posted May 8, 2011 Interesting reading about what might be wrong with Garmin's formulas. Question: If Garmin is using the wrong formula, and getting the wrong sunset time, why is there not the same error for sunrise time? NA0G Quote Link to comment
+StarBrand Posted May 8, 2011 Share Posted May 8, 2011 I still find it overly humorous that so very precious few posts actually mention standing outside and watching the sunset as compared to some much hyped calculation online or in a Garmin unit. Nor do I much care if the calculations differ from the actual sunset moment by +/- a minute or 2. Anybody care to explain a good practical reason why I should care? Quote Link to comment
+Corp Of Discovery Posted May 9, 2011 Share Posted May 9, 2011 I still find it overly humorous that so very precious few posts actually mention standing outside and watching the sunset as compared to some much hyped calculation online or in a Garmin unit. Nor do I much care if the calculations differ from the actual sunset moment by +/- a minute or 2. Anybody care to explain a good practical reason why I should care? Ummmm, in case you're chasing or being chased by vampires? Quote Link to comment
yogazoo Posted May 9, 2011 Share Posted May 9, 2011 (edited) I still find it overly humorous that so very precious few posts actually mention standing outside and watching the sunset as compared to some much hyped calculation online or in a Garmin unit. Nor do I much care if the calculations differ from the actual sunset moment by +/- a minute or 2. Anybody care to explain a good practical reason why I should care? So according to you the fools over at NOAA, Naval Observatory, NASA, etc, are just wasting everyones time with their silly little innacurate sunrise/sunset calculators. Completely pointless, just go outside and look. People who care don't care that you don't care. Look, I agree with you that it's a minor thing. But it's still a mystery that Garmins calculation does not agree with any other calculator. Doesn't it make you the least bit curious why? I'm a photographer in my spare time and like to use the moonrise calculator to grab those photo's of the cool looking rising moon. It's important to me to get a good estimate of what part of the sky the moon will be rising in so I can set up a nice shot with a tripod before it even starts rising. The fact that the Oregon series doesn't tell you is a big deal to me. My point is that some minor feature that doesn't mean squat to you may mean alot to someone who finds a use for it. There's no point to bashing those who find an error in a feature they find value in. If you don't care about it don't post about how much you don't care. Can't we all just get along? Edited May 9, 2011 by yogazoo Quote Link to comment
+janandsteve Posted May 9, 2011 Author Share Posted May 9, 2011 To the doubters: Remember if the sunset calculation is out, so is the fish and hunt calculators...... Which might matter - if there was ANY rational basis in believing the Fish and Hunt Calculator in the first place! I mean - does anyone REALLY think that Alaskan Grizzlies, Indian Tigers, Australian Barramundi, and Scottish Pheasants all follow the same rules in deciding when to come out to feed?! (Or maybe they all carry Garmins in their fur / feathers / fins, and consult the tables to find the best time to forage? ) http://www.arkansasstripers.com/best_times_to_fish.htm Quote Link to comment
+janandsteve Posted May 9, 2011 Author Share Posted May 9, 2011 Interesting reading about what might be wrong with Garmin's formulas. Question: If Garmin is using the wrong formula, and getting the wrong sunset time, why is there not the same error for sunrise time? NA0G Sunset and sunrise formulae are different. The sunrise is also out on the 62s. Quote Link to comment
+janandsteve Posted May 9, 2011 Author Share Posted May 9, 2011 I still find it overly humorous that so very precious few posts actually mention standing outside and watching the sunset as compared to some much hyped calculation online or in a Garmin unit. Nor do I much care if the calculations differ from the actual sunset moment by +/- a minute or 2. One word.... predictions. Anybody care to explain a good practical reason why I should care? Firstly, have a read of this: http://www.arkansasstripers.com/best_times_to_fish.htm Secondly, the 62s is in error in ALL FOUR parameters - sunrise, sunset, moonrise, moonset, which has an effect on the above for those who use that facility..... Thirdly, with Atlantic Bluechart installed, the tide table predictions are also wrong, based on http://www.pol.ac.uk/ With reference to tide tables, heres a typical senario..... A boat moored on a drying out pontoon - that is at low tide, your boat sits on mud. At high tide it floats and can be driven away - Now the calculation of the four main parameters sunrise, sunset, moonrise, moonset - which are WRONG in the 62s - are used to predict the tide times - and if the parameters are wrong, so is the tide data - by anything up to 30mins ( THIRTY ) - and that varies by date. Here in the uk, you can use your boat for prob 10 times a year - you have to be off work, have some spare cash for the expensive fuel, the weather good, the sea state smooth and most importantly the boat needs to be floating on the high tide. You may also want to factor in your friends been available too. Imagine how many times a year all the above conditions are met - prob 10 times a year. With this been the case you need a gps to be accurate and allow you to plan ahead - forget standing outside looking at the sunset - that is just not an issue here. With so little time to enjoy the water, you tend to maximise your time out on the water, hence you need accurate tools to plan the day. Imagine relying on a tool that is WRONG - like the 62s - can you imagine pulling into the harbour to find you pontoon has started to dry out and there is not enough water to land and tie up your boat.... what then, hang around until high tide comes in approx 12 hours? The rate of the outgoing tide means the 30mins deviation from actuals is just not good enough. The chances are you cannot afford to run the boat for 12 hours until the next high water - you tend to maybe use it for just a few hours - hence 30mins deviation is not acceptable. Also, these spur of the moment decisions mean that you do not have access to a pc to do any serious calculations on line - its a bit fo fun on the water not an expedition. Size of the boat and time used does not warrant spending money on a dedicated chartplotter, which accurately predicts the tide. If one manuf can get it right, so should Garmin. The sunrise, sunset, moonrise, moonset times are WRONG. The tide data is WRONG. Garmin know the correct formula - my etrex from 2002 is accurate on sunrise and sunset, so I want my 62s to be just as good - what is wrong with that? Comments about standing outside and comparing the actual sunset to the 62s is ( xxxxxxx ) - fill in your own expletive. I was in 2 minds whether to bother replying to you post, but for the benefit of others, I decided to add the above comments. If you dont have an issue with, or a practical use of the function, then kindly do not post. Those that do care will continue to lobby Garmin until they correct the mistake..... if they cannot get a simple formula right, then it gives me little confidence in their other calculations..... eg the trip odometer is always a lot greater than when you look at the track distance when plotted in pc software - but thats another issue already discussed on the forum. Quote Link to comment
+janandsteve Posted May 9, 2011 Author Share Posted May 9, 2011 I still find it overly humorous that so very precious few posts actually mention standing outside and watching the sunset as compared to some much hyped calculation online or in a Garmin unit. Nor do I much care if the calculations differ from the actual sunset moment by +/- a minute or 2. Anybody care to explain a good practical reason why I should care? So according to you the fools over at NOAA, Naval Observatory, NASA, etc, are just wasting everyones time with their silly little innacurate sunrise/sunset calculators. Completely pointless, just go outside and look. People who care don't care that you don't care. Look, I agree with you that it's a minor thing. But it's still a mystery that Garmins calculation does not agree with any other calculator. Doesn't it make you the least bit curious why? I'm a photographer in my spare time and like to use the moonrise calculator to grab those photo's of the cool looking rising moon. It's important to me to get a good estimate of what part of the sky the moon will be rising in so I can set up a nice shot with a tripod before it even starts rising. The fact that the Oregon series doesn't tell you is a big deal to me. My point is that some minor feature that doesn't mean squat to you may mean alot to someone who finds a use for it. There's no point to bashing those who find an error in a feature they find value in. If you don't care about it don't post about how much you don't care. Can't we all just get along? Good post - Well put! Quote Link to comment
+dfx Posted May 9, 2011 Share Posted May 9, 2011 One word.... predictions. Firstly, have a read of this: http://www.arkansass...mes_to_fish.htm Secondly, the 62s is in error in ALL FOUR parameters - sunrise, sunset, moonrise, moonset, which has an effect on the above for those who use that facility..... Thirdly, with Atlantic Bluechart installed, the tide table predictions are also wrong, based on http://www.pol.ac.uk/ The problem with any and all of those predictions is that the only way to confirm their accuracy is to check their results against real-world data. But how would you ever know at which moment exactly the sun sets? How would you ever know at which moment exactly the tide is the highest? How would anyone ever know? It's not something that can be measured that precisely, the best you can get is a time range and if your prediction gets you a time within that range, then it's good. So there is no "right" or "wrong". All you can ask for (and that's what you're doing) is for Garmin to use the same formula and give the same predictions as what someone else is using. It's not gonna be any more right or wrong. Quote Link to comment
+janandsteve Posted May 9, 2011 Author Share Posted May 9, 2011 One word.... predictions. Firstly, have a read of this: http://www.arkansass...mes_to_fish.htm Secondly, the 62s is in error in ALL FOUR parameters - sunrise, sunset, moonrise, moonset, which has an effect on the above for those who use that facility..... Thirdly, with Atlantic Bluechart installed, the tide table predictions are also wrong, based on http://www.pol.ac.uk/ The problem with any and all of those predictions is that the only way to confirm their accuracy is to check their results against real-world data. But how would you ever know at which moment exactly the sun sets? How would you ever know at which moment exactly the tide is the highest? How would anyone ever know? It's not something that can be measured that precisely, the best you can get is a time range and if your prediction gets you a time within that range, then it's good. So there is no "right" or "wrong". All you can ask for (and that's what you're doing) is for Garmin to use the same formula and give the same predictions as what someone else is using. It's not gonna be any more right or wrong. Yep, I agree, its all about predictions not exact times, but it seems ALL those that do predict the times do so using the same statistical model, and that is all we are asking Garmin to use - so that their units tie in with the online predictions - thats the best we can get. Quote Link to comment
+keehotee Posted May 9, 2011 Share Posted May 9, 2011 Here in the uk, you can use your boat for prob 10 times a year - you have to be off work, have some spare cash for the expensive fuel, the weather good, the sea state smooth and most importantly the boat needs to be floating on the high tide. You may also want to factor in your friends been available too. Imagine how many times a year all the above conditions are met - prob 10 times a year. With this been the case you need a gps to be accurate and allow you to plan ahead - forget standing outside looking at the sunset - that is just not an issue here. With so little time to enjoy the water, you tend to maximise your time out on the water, hence you need accurate tools to plan the day. Imagine relying on a tool that is WRONG - like the 62s - can you imagine pulling into the harbour to find you pontoon has started to dry out and there is not enough water to land and tie up your boat.... what then, hang around until high tide comes in approx 12 hours? The rate of the outgoing tide means the 30mins deviation from actuals is just not good enough. The chances are you cannot afford to run the boat for 12 hours until the next high water - you tend to maybe use it for just a few hours - hence 30mins deviation is not acceptable. Also, these spur of the moment decisions mean that you do not have access to a pc to do any serious calculations on line - its a bit fo fun on the water not an expedition. Size of the boat and time used does not warrant spending money on a dedicated chartplotter, which accurately predicts the tide. If one manuf can get it right, so should Garmin. The sunrise, sunset, moonrise, moonset times are WRONG. The tide data is WRONG. Garmin know the correct formula - my etrex from 2002 is accurate on sunrise and sunset, so I want my 62s to be just as good - what is wrong with that? If my boating was so tide-critical, I wouldn't be relying on a hand-held, consumer grade GPS to dictate when I turned up. Not to mention that I'd have been onboard for at least an hour checking the engines and unpacking the sails before I planned to set off anyway I also make sure I've got a set of good old fashioned hard copy tide tables for my moorings too Quote Link to comment
yogazoo Posted May 9, 2011 Share Posted May 9, 2011 If my boating was so tide-critical, I wouldn't be relying on a hand-held, consumer grade GPS to dictate when I turned up. I don't buy into the, seemingly widespread idea, that just because I ONLY paid $500 for a GPS unit and it's labled as a "Consumer" model, then it can have all kinds of irregularities and I just have to live with it. As far as positioning goes OK, 3 meters, I get it, it's no Trimble. But as far as everything else goes I expect it to work to the highest level of accuracy possible. Quote Link to comment
Now at Zero Gravity Posted May 9, 2011 Share Posted May 9, 2011 Interesting reading about what might be wrong with Garmin's formulas. Question: If Garmin is using the wrong formula, and getting the wrong sunset time, why is there not the same error for sunrise time? NA0G Sunset and sunrise formulae are different. The sunrise is also out on the 62s. But the sunrise error seems much smaller, from other reports. My point: Maybe their algorithm IS the "same old reliable" they've always used but they have a bug in one half of it. With the many complaints of missing features, bugs, and whatnot with the 62, I'd say they brought out the new code for it under considerable time pressure with very little time for fine tuning. They got that Webupdater to fall back on, anyway. Why the heck would they write new code for a feature that has worked for years? Getting the new stuff to work by the release date was hard enough it seems, and continues to be a work in progress... NA0G Quote Link to comment
+StarBrand Posted May 9, 2011 Share Posted May 9, 2011 ... If you dont have an issue with, or a practical use of the function, then kindly do not post. Those that do care will continue to lobby Garmin until they correct the mistake..... if they cannot get a simple formula right, then it gives me little confidence in their other calculations..... eg the trip odometer is always a lot greater than when you look at the track distance when plotted in pc software - but thats another issue already discussed on the forum. ...just asking... I've noticed the 'official' charts for sr/ss times are often off by several minutes one way or the other from actual. I was just kind of wondering if the 62S was somehow a bit closer or further away to actual. Nobody seems to want to discuss that though.....just comparison to some formula. Quote Link to comment
yogazoo Posted May 9, 2011 Share Posted May 9, 2011 (edited) Not just "some formula", THE formula. Somebody better let the scientists over at NOAA and Naval Observatory know that Garmin has a better, more accurate formula with which to estimate SS/SR. That's probably why bugs in their GPS firmware are still worked on years after the units are released. They probably have all their software engineers coming up with better formula's for estimating SS/SR. Sorry, I'm just being a smart-A. I see your point but I still feel Garmin should just make it fall inline with the rest of the world. I guess I don't really care if it's wrong, just as long as it's the standard by which everyone else estimates these times. As of right now, it's not. Now quit poking the badger will ya? Edited May 9, 2011 by yogazoo Quote Link to comment
+julianh Posted May 10, 2011 Share Posted May 10, 2011 To the doubters: Remember if the sunset calculation is out, so is the fish and hunt calculators...... Which might matter - if there was ANY rational basis in believing the Fish and Hunt Calculator in the first place! I mean - does anyone REALLY think that Alaskan Grizzlies, Indian Tigers, Australian Barramundi, and Scottish Pheasants all follow the same rules in deciding when to come out to feed?! (Or maybe they all carry Garmins in their fur / feathers / fins, and consult the tables to find the best time to forage? ) http://www.arkansasstripers.com/best_times_to_fish.htm janandsteve, I am fully aware that there is a so-called "Solunar Theory", and this is programmed into Garmins. My point is - does anyone REALLY belive that ALL different types of (edible) animal, bird and fish use the same "rules" to decide when they should go out and feed? (And by corollary, that's when we should go out hunting and fishing.) Quote Link to comment
+julianh Posted May 10, 2011 Share Posted May 10, 2011 (edited) Thirdly, with Atlantic Bluechart installed, the tide table predictions are also wrong, based on http://www.pol.ac.uk/ janandsteve, I am pretty sure that Garmin marine units do NOT use the algorithm for sunrise / sunset to determine tide times. I would be 99% sure that it uses a completely separate algorithm for tide times, based on "Harmonic Analysis". (e.g. See http://en.wikipedia.org/wiki/Theory_of_tides , and especially, scroll down to the section on "Harmonic Analysis".) The algorithm parameters for various tide stations are included in the BlueChart marine charts that you load onto your GPSr. If you search the various on-line tide table data, you will find a bit of variation on time and height for many tide stations, depending on which source you use. A key determining factor is the number of higher order harmonic terms ("constituents") used, but don't lose track of the fact that it is only an approximation, and other factors such as weather patterns (high and low pressure, prevailing wind direction, etc) can significantly alter the ACTUAL tide height and time of high and low tide at any tide station. Edited May 10, 2011 by julianh Quote Link to comment
+janandsteve Posted May 10, 2011 Author Share Posted May 10, 2011 (edited) Thirdly, with Atlantic Bluechart installed, the tide table predictions are also wrong, based on http://www.pol.ac.uk/ janandsteve, I am pretty sure that Garmin marine units do NOT use the algorithm for sunrise / sunset to determine tide times. I would be 99% sure that it uses a completely separate algorithm for tide times, based on "Harmonic Analysis". (e.g. See http://en.wikipedia.org/wiki/Theory_of_tides , and especially, scroll down to the section on "Harmonic Analysis".) The algorithm parameters for various tide stations are included in the BlueChart marine charts that you load onto your GPSr. If you search the various on-line tide table data, you will find a bit of variation on time and height for many tide stations, depending on which source you use. A key determining factor is the number of higher order harmonic terms ("constituents") used, but don't lose track of the fact that it is only an approximation, and other factors such as weather patterns (high and low pressure, prevailing wind direction, etc) can significantly alter the ACTUAL tide height and time of high and low tide at any tide station. Thanks for your reply and ref to Harmonic Analysis. Laplace Transforms and FFTs went into my brain almost 28yrs ago, so my maths is a little rusty to say the least. The main harmonic constituents still rely on the sun and moon - M2 - Principal lunar semidiurnal constituent S2 - Principal solar semidiurnal constituent N2 - Larger lunar elliptic semidiurnal constituent K1 - Lunar diurnal constituent M4 - Shallow water overtides of principal lunar constituent O1 - Lunar diurnal constituent From what I understand you are saying is that Bluchart uses some sort of table that has been preprogrammed into the software based on the tide station and the above constituents? How big is this table as tidal predictions are possible for years ahead ( 1990 - 2038 )? Does this suggest the unit does some sort of calculations based on the harmonic coefficients to calculate the predicted tidal info? Is the gps current moon and sun info used in these calculations? ( The questions are posed, but I am not necessarily asking you to answer them, if you know what I mean ) All I can say is that the 62s is different with its predictions ( upto 30mins ) compared to a friends Garmin chartplotter ( I dont know the model, but its colour screen and looks expensive ) running Bluechart ( maybe with a different version of software? ) - his chartplotter does tie in with a hard copy of tide tables for the port, to within 2mins. Although many refer to the 62s as 'consumer grade', I would have expected the unit to perform mathematical calculations / table lookups in line with Marine Chartplotters on the market. Are we suggesting Garmin goes to the trouble and alters the calculations for its 'consumer' market? Consumer vs military - I would have expected the accuracy to be better in the military units, but my 'consumer grade' model should be accurate in other calculations. Edited May 10, 2011 by janandsteve Quote Link to comment
+julianh Posted May 11, 2011 Share Posted May 11, 2011 (edited) From what I understand you are saying is that Bluchart uses some sort of table that has been preprogrammed into the software based on the tide station and the above constituents? How big is this table as tidal predictions are possible for years ahead ( 1990 - 2038 )? Does this suggest the unit does some sort of calculations based on the harmonic coefficients to calculate the predicted tidal info? Is the gps current moon and sun info used in these calculations? ( The questions are posed, but I am not necessarily asking you to answer them, if you know what I mean ) janandsteve, I don't have current access to a marine GPSr or marine charts (BlueChart or similar), nor a current set of official tide tables for Australia, so I am working somewhat off memory. However, in the not-too-distant past, I have programmed computer applications to generate tide tables from the official published data using Excel, Mathcad, C_++, etc, so I have some knowledge of the sorts of algorithms that are used. In Australia, the "official" tide charts are the Australian National Tide Tables (ANTT): http://www.hydro.gov.au/prodserv/antt.htm This does not necessarily mean that ANTT is the most sophisticated mathematical tide model around; just that it is the national standard. In the ANTT, as well as the annual day-by-day tabulations of high and low tide predictions, you are given a set of parameters for each port so that you can program your own tide predictions at any time of any day. Some of these parameters are permanent (unchanged from edition to edition), while others are valid for only one year, and are updated in the ANTT each year. I assume this takes into account factors such as additional historical data, more sophisticated analysis of the historical data, morphological changes (tidal effects of recent dredging, seawall construction, etc), and so on. The ANTT provides a couple of different algorithms, which allow you to use more or fewer terms in your predictions. Properly programmed, you will get very similar (but not necessarily identical) predictions as the published tables, depending on which algorithm you use, and how many terms you carry in your calculations. In general, you need to update some of the terms every year, to keep your algorithm totally aligned with the latest copy of the ANTT. If you don't update these parameters every year, you will still get "reasonable" tide predictions, but they will gradually "drift" from the current published ANTT tables. I would suspect that if you buy new BlueChart maps, and compare with older revisions, you will find very similar but not identical tide predictions. I think that ALL of the tide parameters are embedded in the BlueChart data, and probably none are "hard-wired" into the GPSr. You should get identical predictions for any Garmin device running the same version of BlueChart, but you could get different results if you use different versions of BlueChart. For example, you can load the BlueCharts in Garmin MapSource on your PC and get your tide predictions that way - most PCs have no GPS unit or sunrise / sunset function. With respect to why the 62s gets different results - it could be just that a different version of BlueChart is loaded, or it could be a different algorithm (with more or fewer terms, using the same or different tide data model), or it could just be a programming bug - I don't know which! However, 30 minute deviation for the time of high / low tide strongly suggests tide parameters which are badly out-of-date, or else a significant programming bug. Hope this helps! Edited May 11, 2011 by julianh Quote Link to comment
+janandsteve Posted May 11, 2011 Author Share Posted May 11, 2011 From what I understand you are saying is that Bluchart uses some sort of table that has been preprogrammed into the software based on the tide station and the above constituents? How big is this table as tidal predictions are possible for years ahead ( 1990 - 2038 )? Does this suggest the unit does some sort of calculations based on the harmonic coefficients to calculate the predicted tidal info? Is the gps current moon and sun info used in these calculations? ( The questions are posed, but I am not necessarily asking you to answer them, if you know what I mean ) janandsteve, I don't have current access to a marine GPSr or marine charts (BlueChart or similar), nor a current set of official tide tables for Australia, so I am working somewhat off memory. However, in the not-too-distant past, I have programmed computer applications to generate tide tables from the official published data using Excel, Mathcad, C_++, etc, so I have some knowledge of the sorts of algorithms that are used. In Australia, the "official" tide charts are the Australian National Tide Tables (ANTT): http://www.hydro.gov.au/prodserv/antt.htm This does not necessarily mean that ANTT is the most sophisticated mathematical tide model around; just that it is the national standard. In the ANTT, as well as the annual day-by-day tabulations of high and low tide predictions, you are given a set of parameters for each port so that you can program your own tide predictions at any time of any day. Some of these parameters are permanent (unchanged from edition to edition), while others are valid for only one year, and are updated in the ANTT each year. I assume this takes into account factors such as additional historical data, more sophisticated analysis of the historical data, morphological changes (tidal effects of recent dredging, seawall construction, etc), and so on. The ANTT provides a couple of different algorithms, which allow you to use more or fewer terms in your predictions. Properly programmed, you will get very similar (but not necessarily identical) predictions as the published tables, depending on which algorithm you use, and how many terms you carry in your calculations. In general, you need to update some of the terms every year, to keep your algorithm totally aligned with the latest copy of the ANTT. If you don't update these parameters every year, you will still get "reasonable" tide predictions, but they will gradually "drift" from the current published ANTT tables. I would suspect that if you buy new BlueChart maps, and compare with older revisions, you will find very similar but not identical tide predictions. I think that ALL of the tide parameters are embedded in the BlueChart data, and probably none are "hard-wired" into the GPSr. You should get identical predictions for any Garmin device running the same version of BlueChart, but you could get different results if you use different versions of BlueChart. For example, you can load the BlueCharts in Garmin MapSource on your PC and get your tide predictions that way - most PCs have no GPS unit or sunrise / sunset function. With respect to why the 62s gets different results - it could be just that a different version of BlueChart is loaded, or it could be a different algorithm (with more or fewer terms, using the same or different tide data model), or it could just be a programming bug - I don't know which! However, 30 minute deviation for the time of high / low tide strongly suggests tide parameters which are badly out-of-date, or else a significant programming bug. Hope this helps! Hi Julian Thank you for the detailed explanation - much appreciated. I now have a better understanding of how the parameters work. I'm still perplexed to the discrepancy with the 62s.... I have loaded Bluchart into Garmins Homeport software and crossed ref the tide data to http://www.pol.ac.uk/ntslf/tidalp.html. My 'raw' data from Bluechart ties in quite accurately with that site, sugesting that my version of Bluechart is indeed ok with the parameters. Some times are bang on, others are upto 6mins out, but I now understand that this may be due to a different numbers of parameters used - the more parameters used the more 'accurate' the prediction....... but this only adds more fuel to the fire with regards to how the 62s behaves - it looks like it has the right parameters, but does not handle then well.... or maybe only 1 or 2 parameters are downloaded to the 62s from the many used in bluechart on the pc????? Who knows...... maybe because its not a dedicated marine unit - my friends Garmin chartplotter seems fine. Why 'claim' to be an 'all rounder' when the 62s cannot handle it?..... Its good at accuracy ( to 8ft ) and ok with geocaching, but outside of that it is somewhat of a toy. I tried the calculator earlier and that has an error on it too, would you believe..... I typed in 3 x 3 and it gave me the answer of 8.75 - close enough I thought, so no big issue..... ( Joke, before anyone tries to check! ) Maybe Garmin should broaden their horizons to maximise sales and offer this unit for sale in 'Toys-r-Us'... food for thought. Thanks again for the explanation. Steve Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.