Page 1 of 2 12 LastLast
Results 1 to 30 of 49

Thread: Misleading star colors illustration in Sky and Telescope

  1. #1
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265

    Misleading star colors illustration in Sky and Telescope

    I am annoyed by the simplistic color scheme used to illustrate the colors of selected bright stars on the H-R diagram on p. 27 of the latest Sky and Telescope. In my opinion they regressed from the high standards shown by their excellent September 1992 article on star colors. Following is an image I prepared from the two articles.

    Star colors B.jpg

    The upper images are in colors scanned from the 1992 article. They show just how pale the colors are, and they accurately show that the saturation diminishes as we go from both ends toward the center. They are in good agreement with what I understand about incandescent bodies and with what I see with my own eyes when observing these stars. The lower ones, from an H-R diagram they included in the new issue, are excessively saturated in types G through M. They do not show the reduced saturation in K and G, and they give the impression that the color stays the same throughout K and abruptly changes when going into type G. They inexplicably made F almost identical to A. This is typical of illustrations throughout most of the 20th century, in which the authors apparently took the jargon commonly used by astronomers too literally. The late S&T editor Joseph Ashbrook, a stickler for technical accuracy, must be turning over in his grave. The 1992 article was consistent with the standards he had set before his death in 1980.

  2. #2
    Join Date
    Aug 2006
    Posts
    3,164
    I hope you wrote to them and sent this verbatim.

  3. #3
    Join Date
    Sep 2003
    Posts
    12,396
    Quote Originally Posted by Hornblower View Post
    I am annoyed by the simplistic color scheme used to illustrate the colors of selected bright stars on the H-R diagram on p. 27 of the latest Sky and Telescope. In my opinion they regressed from the high standards shown by their excellent September 1992 article on star colors. Following is an image I prepared from the two articles.

    Star colors B.jpg

    The upper images are in colors scanned from the 1992 article. They show just how pale the colors are, and they accurately show that the saturation diminishes as we go from both ends toward the center. They are in good agreement with what I understand about incandescent bodies and with what I see with my own eyes when observing these stars. The lower ones, from an H-R diagram they included in the new issue, are excessively saturated in types G through M. They do not show the reduced saturation in K and G, and they give the impression that the color stays the same throughout K and abruptly changes when going into type G. They inexplicably made F almost identical to A. This is typical of illustrations throughout most of the 20th century, in which the authors apparently took the jargon commonly used by astronomers too literally. The late S&T editor Joseph Ashbrook, a stickler for technical accuracy, must be turning over in his grave. The 1992 article was consistent with the standards he had set before his death in 1980.
    Yep, that lower color depiction is horrible. Are they perhaps using them simply as a correlation with the spectral color scheme found in the H-R diagram, thus not what we actually observe? I realize this is unlikely, but nothing else makes sense. Do they at least mention or imply that these are colors only as seen below the atmosphere, and not as seen in space (attenuated) without extinctions that produce excess yellow, orange, and red coloring?

    Once again, Capella rises up as the yellow example. It's the yellow pigment stigmant that led to the Sun being cast into the yellow "type" of star. It's spectrum was seen by Fr. Secchi as matching close enough to the Sun to group the Sun with these yellowish stars, and perhaps, the Sun was yellow ever since, [I think his contemporary Danti may have preceded this yellow classification (soon to be part of the Secchi Type)] So I've wondered that if you make the G class yellow, then is it just too easy to make K stars orange and M stars red, ignoring hard ground-based observational data that would bother to adjust for varying particulate counts that effect the extinction levels?
    Last edited by George; 2018-Oct-24 at 03:49 PM.
    We know time flies, we just can't see its wings.

  4. #4
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    Philip Steffey, Ph.D., the author of the 1992 article, said that the star colors as displayed on his chart are as seen from sea level at an elevation of 45 degrees. He described stars in the range A5-F5 as looking neutral white, with earlier types having increasing amounts of blue and later types having increasing amounts of yellow. He described Deneb (A2) and Fomalhaut (A3) as cold white, with Vega (A0) noticeably bluish white. Early B stars such as Spica show somewhat more blue, while the hotter O stars actually appear slightly less blue than Spica because of greater departures from blackbody tint as a result of absorption in their extended atmospheres. Late F stars such as Polaris (F8) tend to be warm white, while the G stars are very pale yellowish white. My conclusion from various sources about atmospheric discoloration is that the G stars would look neutral white from out in space. My hunch is that Father Secchi saw Capella much as I did before my cataract surgery. It looks paler now, though still off-white, while the A and B stars look slightly bluer than before.

    For reasons that seem arbitrary to me, professional astronomers a century ago chose to designate Vega as neutral white, regardless of the bluish tint most of us see. Then they used color names without pastel-indicating adjectives as jargon, which popular media writers took too literally and published in a self-perpetuating pattern. In accordance with the high standards set by founder Charles Federer and maintained by Mr. Ashbrook, Dr. Steffey broke with that practice and gave us a realistic presentation. In my opinion the present management was remiss in admitting a regression to the old ways. I wrote a letter to the editor saying pretty much what I did in my opening post, but I toned it down by not including a remark about Mr. Ashbrook turning over in his grave. I am curious to see if and how they respond to my objective remarks.

  5. #5
    Join Date
    Sep 2003
    Posts
    12,396
    Very nice post, Hornblower! I hope you won't mind me parsing it as it contains a number of interesting and helpful nuggets.

    Quote Originally Posted by Hornblower View Post
    Philip Steffey, Ph.D., the author of the 1992 article, said that the star colors as displayed on his chart are as seen from sea level at an elevation of 45 degrees.
    Ok, that's a logical observing basis. It's an air mass of 1.4, ignoring any unusual levels of particulates. This gives some reddening but not that much. The Sun will still project as a white star (i.e. no hint of yellow except perhaps at the limb for several reasons, perhaps).

    He described stars in the range A5-F5 as looking neutral white, with earlier types having increasing amounts of blue and later types having increasing amounts of yellow. He described Deneb (A2) and Fomalhaut (A3) as cold white, with Vega (A0) noticeably bluish white. Early B stars such as Spica show somewhat more blue, while the hotter O stars actually appear slightly less blue than Spica because of greater departures from blackbody tint as a result of absorption in their extended atmospheres.
    That's interesting that surface extinctions cause some desaturation for O stars. A pure BB profile for the hottest stars reveal that it's hard to get the blue-end to jump-up enough to produce a saturated blue, so even a perfect BB won't look blue at even the highest of star temperatures, millions of degrees are required for this.

    This helps explain, perhaps, why Secchi's initial star types (1862) started with white stars (no blue stars), which included Vega. He added a blue star class in 1867, interestingly, which included Vega. But by the end of that year, he recognized the importance of the spectral lines as being superior for classification. So, IMO, the importance of getting colors correct became far less, thus easier to allow misapplication of star color assignments, as in the HR diagram (assuming we are taking those colors as actual star colors and not just helpful to accurate spectral assignments). This hiccup would get compounded given that all the advanced spectrographic work was black and white (photography).

    Late F stars such as Polaris (F8) tend to be warm white, while the G stars are very pale yellowish white.
    I suspect that the hotter G stars, G2 in particular, will require a little extra in particulate extinctions. Solar twins look and image white, though I don't have any hard viewing of them from sea level, admittedly. However, after all the extensive effort (albeit clumsy) given to heliochromology demonstrably falsifying any claim that the Sun (as seen above our atm.) to have any yellow tint, I observed eta Cas. (B-V = 0.58) through an 8" Dob. reflector and it had a beautifully distinct gold-yellow tint to it, though it should have been more white than any solar twin. It was enjoyable to look at especially since its red dwarf companion was a brilliant red. This took place at a higher elevation (McD obs.) than normal and only a little less than 45 deg. in alt., IIRC! I have since observed it from home ( ~ 700' elev.) and each of the few times I saw it, it was white. Also, the red dwarf was more orange than red, further supporting the particulate issue. I learned that particle counts at the observatory are measured and, in some cases, the domes are, on occasion, shut-down due to excessive levels. I feel sure that I picked a bad night to see eta Cas. Perhaps it doesn't take much in the way of aersoils, pollens, etc. to affect a yellowish result for G class stars. Secchi may have had this issue, plus the fact he was using a refractor (Merz lens, however) that likely exhibited chromatic aberration.

    Further observations would better reveal if I'm right or not but I lost interest once the scales fell off my eyes when seeing the white solar projection at Kitt Peak; this killed any hint of yellow for the Sun as a star (AM0 of course).

    My conclusion from various sources about atmospheric discoloration is that the G stars would look neutral white from out in space.
    Can we not say this is incontrovertible (at least for the hotter G stars)? If a quality, unfiltered projection of the Sun is pure white (center to limb!) as seen anywhere terrestrially (e.g. Kitt Peak's McMath-Pierce projection room) then what atmospheric extinctions could possible be added back into it to produce a yellow tint? These extinctions are blues and some greens, so the only possible color change would be to see a tiny hint of blue and likely only in the central zone where the photosphere temperature is about 6390K vs. 5000K for the limb. Nevertheless, it is unlikely that any blue would be seen since the photon flux density is still very flat across the visible spectrum in these temperature ranges.

    My hunch is that Father Secchi saw Capella much as I did before my cataract surgery. It looks paler now, though still off-white, while the A and B stars look slightly bluer than before.
    Yes, and I suspect his use of a refractor may have increased this effect. Also, I am fairly sure that Danti already saw and assigned yellow to Capella, which may or may not have contributed, in some small bias way. It would be nice to know what particulate issues may have affected their work. It's, at best I suppose, curious that Schiaperelli admitted to a mild case of Daltonism when he discussed the colors he saw for Mars. My guess is that Fr. Secchi would not be affected by other's color opinions but that conditions for Rome and his optics would be key factors.

    For reasons that seem arbitrary to me, professional astronomers a century ago chose to designate Vega as neutral white, regardless of the bluish tint most of us see.
    Vega seems to be an interesting case even if we only look at Secchi's work. This is another reason why I am curious about all the extinction factors Italian astronomers faced in the latter 19th century.

    Then they used color names without pastel-indicating adjectives as jargon, which popular media writers took too literally and published in a self-perpetuating pattern.
    Yes, this seems to be the case. It is clear in reading about star colors from that time frame that they were quick to recognize that it was subjective given that not all eyes see things the same as others. IIRC, Schiaparelli was given as an example of this. Color designations, nevertheless, is an effective way to classify things in the sense of memorizing and recognizing, at the expense of accuracy. With spectral classification schemes being introduced by folks like Murray and Cannon, using B&W spectrographs, color may have been seen as a helpful secondary tool to get spectral classes established since it is, well, colorful. Things we enjoy will be "self-perpetuating", as you say.

    In accordance with the high standards set by founder Charles Federer and maintained by Mr. Ashbrook, Dr. Steffey broke with that practice and gave us a realistic presentation. In my opinion the present management was remiss in admitting a regression to the old ways. I wrote a letter to the editor saying pretty much what I did in my opening post, but I toned it down by not including a remark about Mr. Ashbrook turning over in his grave. I am curious to see if and how they respond to my objective remarks.
    Your point is sure encouraging for me to read-up on these astronomers to better understand star color and the errors that come from not taking color accuracy a little more seriously, and a little less artistically.
    Last edited by George; 2018-Oct-25 at 04:50 PM. Reason: grammar
    We know time flies, we just can't see its wings.

  6. #6
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    Here is a cleaned-up and brightened version of my illustration. I made the bluest one as bright as possible and then adjusted the others to about the same luminance, so the blue ones would not be fainter than the neutral one as is too often the case in printed media.
    Star colors D.jpg
    For magazine and textbook prints I would not object to boosting the saturation provided we say it is boosted and why. The reason would be to prevent discoloration in the paper and the printing process from masking subtle differences in pastel tints. Here is a boosted version.
    Star colors C more saturated.jpg
    In both cases only the upper row, from the 1992 article, is adjusted from my earlier post.

    If I am not mistaken, the projection of the Sun onto that screen at McMath is plenty bright to go neutral white in our perception of it. My projections in a blacked out room certainly look neutral. Only with stars at much fainter levels do I see a slight yellowish tint. Bright early G stars on the main sequence are rare birds, but any star with the same color index will do. My research indicates that Polaris is close enough. It is about midway between Procyon and Capella.

  7. #7
    Join Date
    Sep 2003
    Posts
    12,396
    Quote Originally Posted by Hornblower View Post
    Here is a cleaned-up and brightened version of my illustration. I made the bluest one as bright as possible and then adjusted the others to about the same luminance, so the blue ones would not be fainter than the neutral one as is too often the case in printed media.
    Star colors D.jpg
    For magazine and textbook prints I would not object to boosting the saturation provided we say it is boosted and why. The reason would be to prevent discoloration in the paper and the printing process from masking subtle differences in pastel tints. Here is a boosted version.
    Star colors C more saturated.jpg
    In both cases only the upper row, from the 1992 article, is adjusted from my earlier post.
    Something like this should be adopted as a standard for all respectable work. But...

    If I am not mistaken, the projection of the Sun onto that screen at McMath is plenty bright to go neutral white in our perception of it.
    It's actually not that bright and it's fairly bright in the room. I think there are windows, so daylight is the higher temperature light source, and I think I recall fluorescent lighting as well, which I would question as having an effect on the color determination, if it weren't for the fact that our brains have a powerful color constancy control that will make something semi-white look white, as I understand it.

    My projections in a blacked out room certainly look neutral.
    The typical projection I see in ambient lighting is only of a white disk. Since the blue sky will surround the image, then if I understand contrasting color effects, it should induce a yellow tinge, but it doesn't, or at least I think this is the case. I have a Sunspotter that also shows only a white disk when the Sun is at a reasonable altitude. Of course, at lower and lower altitudes yellow advances from the limb toward the center.

    Would you agree the Sun should be labeled white if you found that all solar projections were produced only white disks? [Ignoring low altitude circumstances.]

    Procyon color claims from the web seem to vary from a tinge of yellow to white. I wonder if the yellow tint might be due to it possibly being bright enough on the fovea -- though it's only about a 0 mag. -- and, as a point-source, having a yellowish result due to the lack of blue color cones in the fovea?

    Only with stars at much fainter levels do I see a slight yellowish tint.
    Would the above help explain this?
    Last edited by George; 2018-Oct-26 at 02:50 PM.
    We know time flies, we just can't see its wings.

  8. #8
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    1,400
    Quote Originally Posted by Hornblower View Post
    Here is a cleaned-up and brightened version of my illustration. I made the bluest one as bright as possible and then adjusted the others to about the same luminance, so the blue ones would not be fainter than the neutral one as is too often the case in printed media.Star colors D.jpg
    Downloaded this image in MS Paint, clicked on "Mid G: Capella" in the top row, then got a color reading. It was:

    Hue: 37
    Sat: 97
    Lum: 196

    Red: 227
    Green: 224
    Blue: 189

    Now everyone knows exactly what color that was, and they can duplicate it. (There are better ways to do this, this just looked simple.) This information would have to go out to printers, art directors, editors, etc. if this is the color that is needed.
    There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.
    — Mark Twain, Life on the Mississippi (1883)

  9. #9
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    Quote Originally Posted by Roger E. Moore View Post
    Downloaded this image in MS Paint, clicked on "Mid G: Capella" in the top row, then got a color reading. It was:

    Hue: 37
    Sat: 97
    Lum: 196

    Red: 227
    Green: 224
    Blue: 189

    Now everyone knows exactly what color that was, and they can duplicate it. (There are better ways to do this, this just looked simple.) This information would have to go out to printers, art directors, editors, etc. if this is the color that is needed.
    That's the same RGB formula as in my PaintShop Pro, in which I did the artwork. It appears to be a standard system for use with color monitors.

  10. #10
    Join Date
    Sep 2003
    Posts
    12,396
    I got my S&T mag. just yesterday and read the article. You could have hit them harder given some of the other ambiguity within their presentation:

    1) Their opening sentence starts off appropriately by stating the H-R diagram is a plot of luminosity and spectral classes, but they drift away from this by ending on color and size.
    2) The statement they make abut the Sun, "You can immediately tell that it is a yellow star of moderate temperature..." is only true because their HR diagram has colored the Sun red, but not any of the stars around it. Thus, there is no way to immediately tell that it is a yellow star, unlike the normal, colorful H-R diagrams. It is also hard to tell with their diagram if it is a G2 or G3 or even G4 since no vertical lines or a grid is presented.
    3) Given the proximity of Capella and Pollux, it's hard to imagine that anything close to that level of saturation could be possible. Ignoring a normal color hue transition with slight spectral class differences is illogical, yet used consistently and erroneously in the presentation.
    4) The claim the Luminosity class for the "V" stars is assigned to dwarfs, yet they fail to mention this is for main sequence dwarfs only. White dwarfs, for example, are not in this class.
    5) They, perhaps wisely, do not show any colors for the white dwarfs.
    We know time flies, we just can't see its wings.

  11. #11
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    1,400
    Quote Originally Posted by Hornblower View Post
    That's the same RGB formula as in my PaintShop Pro, in which I did the artwork. It appears to be a standard system for use with color monitors.
    Good. Having those details on hand for reprinting the images would (probably not) prevent someone from guessing at the color and getting it wrong, but at least you tried.
    There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.
    — Mark Twain, Life on the Mississippi (1883)

  12. #12
    Join Date
    Aug 2005
    Location
    NEOTP Atlanta, GA
    Posts
    2,374
    Quote Originally Posted by Hornblower View Post
    That's the same RGB formula as in my PaintShop Pro, in which I did the artwork. It appears to be a standard system for use with color monitors.
    Yes, monitors. But not printing, which uses the CMYK color space.

    Achieving fidelity between the two color spaces is a challenge that dates back to the earliest days of digital graphics. Most commercial printers have RIPs that can do a fair job of matching but sometimes the bit gets dropped, so to speak.


    Sent from my iPhone using Tapatalk Pro

  13. #13
    Join Date
    Sep 2003
    Posts
    12,396
    It's fun noting that Hornblower's RGB values are somewhat close in a correlation with a BB photon flux distribution for the typical (IMO) red, green and blue wavelengths.

    The RGB result (using 750nm, 530nm, 470nm) equates to (normalizing in red):
    227
    212
    187

    [Added: The reduced green value makes for a noticeable hue difference, but this is only a BB approach.]

    It would be more interesting to see how this looks with an actual SED for Capella. The Sun is about 10% higher in actual photon flux in blue (~5% for green) than its 5850K BB profile, and Capella's blue and green value would come up as well. I don't doubt an equation could be made to produce RGB values given any SED for a main sequence star, perhaps some giants as well (like Capella), but it might look pretty ugly.
    Last edited by George; 2018-Oct-26 at 07:05 PM.
    We know time flies, we just can't see its wings.

  14. #14
    Join Date
    Sep 2003
    Posts
    12,396
    Quote Originally Posted by schlaugh View Post
    Yes, monitors. But not printing, which uses the CMYK color space.

    Achieving fidelity between the two color spaces is a challenge that dates back to the earliest days of digital graphics. Most commercial printers have RIPs that can do a fair job of matching but sometimes the bit gets dropped, so to speak.
    I'm glad you mentioned this as I had some artwork in CMYK where the files were too large to email, but I noticed if I converted the original to an RGB file, the file size was significantly reduced. It seems to me the gain in file size, at least for these, outweighs what slight hue changes might take place in printing. Is this fair?
    We know time flies, we just can't see its wings.

  15. #15
    Join Date
    Aug 2005
    Location
    NEOTP Atlanta, GA
    Posts
    2,374
    Like many things it’s a trade off and I’m not sure how best to answer your question. Maybe to ask what is the end goal; viewing on screen or in print? If screen only then the file size reduction may be OK.

    S&T of course - I hope - wants to align the star colors to observations and accepted definitions supported by contemporary data. (And FWIW this thread is personally edifying as to how stellar color and luminosity evolved. I had not heard of Fr Secchi before now. )

    When I read the OP my thought was “oh, offset printing vs. digital printing strikes again.” I don’t know this but I suspect the discrepancy between 1992 and today was caused by a change in printing technologies.

    Editorial malfeasance comes in to play because apparently no one cared to create a more faithful match to the original article. Could it be plain ignorance? I’m very curious to any response from S&T.



    Sent from my iPhone using Tapatalk Pro

  16. #16
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    1,400
    Quote Originally Posted by schlaugh View Post
    Editorial malfeasance comes in to play because apparently no one cared to create a more faithful match to the original article. Could it be plain ignorance? I’m very curious to any response from S&T.
    Former magazine editor here, with 17.5 years experience -- yes, thank you, exactly. A lot of supposedly responsible people will settle for "sort of like it" instead of "exactly like it". You got it right.
    There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.
    — Mark Twain, Life on the Mississippi (1883)

  17. #17
    Join Date
    Jul 2005
    Posts
    16,788
    Transferring colours between colour spaces isn't as much of a black art as it once was, speaking as someone who nowadays regularly sends an image composed on the computer monitor to an 8-colour printer, and then holds the result print next to the monitor to assess the match. It just requires a smidgeon of knowledge, some patience, and a bit of care.
    I would guess in magazine production there are commercial limitations arising from paper quality, printing speed, and the trade-off between colour gamut and the cost of additional inks, but these should be well-understood by any publisher involved in reproducing astronomical images. I guess the reproduction of images provided by contributors who haven't calibrated their monitors properly might also be a problem.

    This discussion reminds me that I once sold an article about star colours to Astronomy magazine, back in the 80s before it turned into a comic. When the piece hadn't appeared in print for three or four years, I dropped them a letter asking what had happened. They informed me that editorial policy had changed and they would no longer be using the article, formally returned the rights to me, and told me to keep the fee. I never got around to submitting it anywhere else, though.

    Grant Hutchison
    Blog

    Note:
    During life, we all develop attitudes and strategies to make our interactions with others more pleasant and useful. If I mention mine here, those comments can apply only to myself, my experiences and my situation. Such remarks cannot and should not be construed as dismissing, denigrating, devaluing or criticizing any different attitudes and strategies that other people have evolved as a result of their different situation and different experiences.

  18. #18
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    I don't know how to convert between RGB and CMYK, but it appears that my computer does a good job of it, because my printers make color prints that match what is on the monitor as well as can be reasonably expected from the limitations of the printing process. I heartily concur with an inference of editorial malfeasance. They had a realistic set of colors and Dr. Steffey's technical writing about it in their archives, and they either forgot about it or chose to disregard it in favor of the simplistic garbage of the past, for unknown motives.

    If I had written about all of the faults I found in the text on p. 27, my letter to the editor probably would have had about 1,000 words rather than the 130 or so I actually sent. The magazine asks us to limit this to 250 words or less, so I kept it brief by limiting my remarks to the colors. I don't mind digressing into these other faults in this thread.

    By the way, George, where did you see a red Sun? On my copy the Sun is practically the same shade of yellow as Capella.
    Last edited by Hornblower; 2018-Oct-26 at 09:59 PM. Reason: Fix punctuation

  19. #19
    Join Date
    Sep 2003
    Posts
    12,396
    Quote Originally Posted by schlaugh View Post
    Like many things it’s a trade off and I’m not sure how best to answer your question. Maybe to ask what is the end goal; viewing on screen or in print? If screen only then the file size reduction may be OK.
    The main goal was to be able to email the graphics and I had to use RGB conversion to accomplish this. I think all the printed work was fine else I would have noted it. Very slight color variations would not likely be apparent, I suppose, since the art was mostly regarding machinery. I did compare both files with appropriate scrutiny, no doubt, (CMYK and RGB) but only on my screen, hence the reason I ask. I have not bothered to compare the print but, as noted, I haven't yet seen any issues.

    S&T of course - I hope - wants to align the star colors to observations and accepted definitions supported by contemporary data.
    Now that I've seen the article, it looks like something done more on a general presentation level than what it should have been for S&T. A simple sentence or two should have been given to clarify their disinterest in color correctness so that the reader might recognize that color accuracy was not their intent.

    (And FWIW this thread is personally edifying as to how stellar color and luminosity evolved. I had not heard of Fr Secchi before now. )
    He is arguably two fathers: Father Secchi and father of spectroscopy. With over 300 stars classified, his three (later up to five) "types" of star were highly respected as a classification system and his time in the U.S. only helped gain this respect. He hand drew in color stellar spectrums along with their "dark rays" and soon realized that the lines were likely the more important aspect to consider for any stellar classification system, though color also correlated with spectrums as well. Soon with photography, these color drawings ended, and with it, the colorful touch associated with spectrums, at least I think this may have contributed to the lack of stellar color associations since the spectral lines were far more objective in any classification system.

    Editorial malfeasance comes in to play because apparently no one cared to create a more faithful match to the original article. Could it be plain ignorance? I’m very curious to any response from S&T.
    The coloring is both overly saturated and simplistic, and the fact that they never suggest otherwise tells me they weren't trying to get star colors correct. It's a very short article (less than 1/2 page) that is entitled, "Where do our bright stars sit in the Hertzsprung-Russell diagram?" Their star colors give general placement on the H-R diagram. However, since they show a table that, erroneously, has only the color "characteristic" associated with the class, then they are misrepresenting what we see for stars. M stars are rarely red, O stars are not blue as shown it the H-R graph where Rigel is the same color as Spica.
    Last edited by George; 2018-Oct-26 at 10:06 PM.
    We know time flies, we just can't see its wings.

  20. #20
    Join Date
    Aug 2005
    Location
    NEOTP Atlanta, GA
    Posts
    2,374
    And I have misused the word “malfeasance “. I think a better word is simply “neglect”.


    Sent from my iPhone using Tapatalk Pro

  21. #21
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    I just now made a test print of the pastel version on plain paper with my inexpensive Hewlett-Packard all-in-one machine. All of the colors are in excellent agreement with what I see on the monitor. S&T printed just such a set of color swatches without a hitch 26 years ago, and the commercial printing presses have continued to improve. I stand by my opinion that the current illustration is a result of editorial neglect, not of technical difficulties in production.

  22. #22
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    A recurring question is about the true color of the Sun. Any answer depends on how we define neutral white for emitters and what is useful as a primary standard for spectral photometry. We can have a wide range of broad spectrum emitters we can call neutral white depending on the task at hand, since they all look colorless when they dominate our ambient light. From my own observations of stars and from what I have learned about the discoloring effect of our atmosphere, I am pretty certain that a star the same color as the Sun would look neutral white from out in space, so for astronomical purposes I am fine with saying the Sun's color is neutral white by definition. When looking at stars in a dark background it appears that my daytime color adaptation goes away and I see other colors for cooler or hotter stars.

    Now let us discuss the prevailing astronomical jargon. It appears that type A0 main sequence stars such as Vega were chosen as a primary standard for determining the color index of a star. Vega is easy to observe, has the least clutter in its spectrum, and is at the temperature where the Balmer lines of hydrogen are strongest. For spectral photometric purposes that makes it a good choice for a primary standard. What is not astronomically necessary, in my opinion, is calling it neutral white when most observers see a distinct trace of pale blue unless the air is severely polluted with brown smog or something similar. If the air over Rome was as murky as was commonplace in places like London in that era, perhaps Father Secchi really did see neutral white. In any case a respected pioneer in stellar spectroscopy attached one-word color terms to various stars and the jargon stuck. It is easier to say things like red, yellow, white or blue than to recite the objective mathematical terminology.

  23. #23
    Join Date
    Sep 2003
    Posts
    12,396
    Quote Originally Posted by Hornblower View Post
    A recurring question is about the true color of the Sun. Any answer depends on how we define neutral white for emitters and what is useful as a primary standard for spectral photometry. We can have a wide range of broad spectrum emitters we can call neutral white depending on the task at hand, since they all look colorless when they dominate our ambient light. From my own observations of stars and from what I have learned about the discoloring effect of our atmosphere, I am pretty certain that a star the same color as the Sun would look neutral white from out in space, so for astronomical purposes I am fine with saying the Sun's color is neutral white by definition.
    I agree but we can demonstrate that it is less subjective than most people, including scientists, think.

    Galileo did not prove the Copernican model when his telescope revealed that Venus has both crescent and gibbous phases, but it did the one thing every proper scientific hypothesis and theory is vulnerable to -- falsification. He falsified the Aristotle/Thomist/Ptolemy theory. It is a valid hypothesis to claim the true color of the Sun (as we would see it with neutral attenuation from space) is yellowish-white, or any other tint from white. After multiple efforts to disprove this, falsifying a yellow claim was something so simple that I all but lost interest in the entire topic. My avatar is, like Galileo's dual phases of Venus evidence, debunks any yellow tint claim for the Sun. It is an unfiltered projection of the Sun from the projection room of the world's largest solar telescope.

    It's funny how I spent the entire night at KP (Kitt Peak)-- not to mention it was our 25th anniversary night (thanks Hun!) -- on a KP 16" RC scope trying to get solar twins like 18 Sco imaged with special techniques to tickle-out its true color, then to get a tour of the McMath-Pierce solar observatory without understanding what I had just observed. Perhaps I became more awake as we drove off when, all of sudden, it hit me like a ton of bricks when I realized that their white image totally destroys any claim for a yellow or yellowish Sun.

    Dr. Potter and Roy Lorenz took this image for me, adding the RGB color pieces they scavenged-up to accommodate me to demonstrate it was a color image. I later bought and sent them a photometer to measure its intensity to demonstrate it is well within the normal photopic vision range. [Hornblower wisely questioned this earlier and I meant to mention this. Objects overly bright will disrupt true color determination.]

    There were many who questioned Galileo's wimpy telescope and the bold claims against a 2000 year theory, which got fused into their theology since Thomas Aquinas' work. His observations were not confirmed by all and there must have been good reasons why he saw things falsely. [The Jesuits, apparently, used a telescope and were fairly quick to confirm Galileo.]

    So, if you, like most, take the Sun's color as something more of a subjective-type claim and not one that important to scientific progress -- it isn't really that important, admittedly -- then you are missing a moment to appreciate the scientific method at work. Please look at my avatar, think about the circumstances, and see how it destroys a yellowish claim. Ambient lighting must be considered, admittedly, but, like a fuzzy crescent Venus, such issues are easily resolved, IMO.

    When looking at stars in a dark background it appears that my daytime color adaptation goes away and I see other colors for cooler or hotter stars.
    Yes, though a black background should be the standard when stating a star's true color. But you will find that images of daytime solar projections will have a white Sun surrounded by a blue tint background due to the fact the the sky is blue and that is the color being reflected in the shadow of the telescope doing the projection. This blue background will tend to make some objects appear more yellow than normal due to the way the eye behaves, further arguing for a white Sun. [Example below from Purdue Engr.]

    PurvesCube.jpeg

    A contrary argument could be that a blue background in solar projection images (just Google them and there are probably hundreds) is more blue than normal because the camera color processing would take the brightest source and shift it, assuming it had a hint of yellow, to it being a white source, thus making it white. Cameras do this for a reason: that's what our brains do, so they are trying to represent what we see as color, and what we see is the definition of color; all camera-oriented color computer models strive to replicate what most of us already have.

    If we can anchor the Sun as a white star, then it should help to open the minds of those who are misusing (unwittingly no doubt) the secondary color scheme found in the H-R diagram.

    Now let us discuss the prevailing astronomical jargon. It appears that type A0 main sequence stars such as Vega were chosen as a primary standard for determining the color index of a star.
    I'm guessing the history behind this goes back to prominent leaders like Secchi.

    Vega is easy to observe, has the least clutter in its spectrum, and is at the temperature where the Balmer lines of hydrogen are strongest. For spectral photometric purposes that makes it a good choice for a primary standard. What is not astronomically necessary, in my opinion, is calling it neutral white when most observers see a distinct trace of pale blue unless the air is severely polluted with brown smog or something similar. If the air over Rome was as murky as was commonplace in places like London in that era, perhaps Father Secchi really did see neutral white.
    That's an important question and the answer is he did not.

    Le Soleil (The Sun) was perhaps the most significant book by Secchi. Unfortunately, I cannot find an English translation of this book but I did manage to translate his table of contents and found, in chapter 7, that he addressed his star types, which grouped primarily by spectral lines and secondarily by color (so I think this is the case. Anyone hear read French?) He admits that what we really see for Vega includes a hint of blue though he put it with white stars, though this may have been when he just had three types. He later included blue-white stars and I'm just being a little too rushed to dig-up the info on it right now.

    In any case a respected pioneer in stellar spectroscopy attached one-word color terms to various stars and the jargon stuck. It is easier to say things like red, yellow, white or blue than to recite the objective mathematical terminology.
    Yep, I think that's hitting the nail on the head. Color gives the 30,000 ft. view of things when we are now in a world of 10' observations that interest us. Once the importance of the spectral lines became understood AND, almost simultaneously, black and white spectrographs ruled the day, color became just another enjoyable and artistic touch to things. Nevertheless, S&T should have stayed at least within the 30,000 ft. range and not jump to over 300,000 ft. without telling anyone their altitude.
    Last edited by George; 2018-Oct-28 at 07:31 PM.
    We know time flies, we just can't see its wings.

  24. #24
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    I just now recovered my observations from just before my first cataract surgery, over three years ago.

    Star B-V

    Epsilon Ori -0.19 pale bluish white
    Sirius 0.01 cold white
    Procyon 0.42 warm white
    Alpha Per 0.48 warm white
    Polaris 0.60 pale yellowish white
    Capella 0.80 pale yellow
    Betelgeuse 1.85 pale orange

    My notes from a couple of weeks later, after my right eye was sufficiently healed for a good look, indicate that a neutral star looked like type G in my left eye and type B in my right eye. With both eyes in binoculars my sensation was pretty much as it was just before the surgery. It appears that my brain cells were splitting the difference. It will be interesting to re-observe these same stars now and see where my neutral white point is. I am curious to see whether or not my adjustment to the major change in what reaches my retinas in daylight has carried over to the set point I have when looking at stars. I have looked at Polaris plenty of times when checking the collimation of my big Dob, but I did not pay much attention to the color. Wouldn't you know, it's cloudy now, so this will have to wait.

    It would be interesting to have numerous observers look at simulated stars in a dark room under laboratory conditions, over a range of color index in increments of 0.1 magnitude, and find the median of what they report as neutral white.

  25. #25
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    The sky cleared as I was writing my previous post, so I took a look. There is little if any change from what I recorded in my old notes, so clearly the recalibration my brain did with the change in the daylight throughput carried over to my perception of star colors at night. Of course I could not get to all of those stars, but I got Polaris and Capella, along with Deneb, Vega and Altair.

    Polaris - warm white to pale yellowish white
    Capella - pale yellowish white
    Altair - neutral white, agrees with Dr. Steffey
    Deneb - cold white
    Vega - pale bluish white

    These are a far cry from the strong shift toward blue I originally had in my right eye, along with a shift toward red in my left eye, immediately after the surgery on the right eye. It appears that my visual cortex went from being tuned for the previously impaired view in both eyes to the temporary mixed messages and on to the fully stabilized state with both eyes fixed.

    I stand by by inference that the Sun, moved out to a few lightyears and viewed from above the atmosphere, would look neutral white and thus would be a suitable standard for defining a neutral white emitter.

  26. #26
    Join Date
    Sep 2003
    Posts
    12,396
    Quote Originally Posted by Hornblower View Post
    The sky cleared as I was writing my previous post, so I took a look. There is little if any change from what I recorded in my old notes, so clearly the recalibration my brain did with the change in the daylight throughput carried over to my perception of star colors at night.
    That's interesting and it's nice to see the account of this given how the blue background could be argued to have a yellowing effect on the star. There are arguments as such as to why the Sun often has a tint of yellow during a very quick glancing view of it high in the sky. I'm guessing the power of color constancy is more remarkable than we guess.

    Of course I could not get to all of those stars, but I got Polaris and Capella, along with Deneb, Vega and Altair.

    Polaris - warm white to pale yellowish white
    Capella - pale yellowish white
    Altair - neutral white, agrees with Dr. Steffey
    Deneb - cold white
    Vega - pale bluish white
    Nice report. I notice you have a tint of blue for Vega but not for Sirius though they have the same B-V value. I will guess that it's due to the large altitude difference during the daytime; Vega being very close to zenith, I think, at one point. Is this likely or do you think the daytime vs. nighttime viewing is the better account?

    These are a far cry from the strong shift toward blue I originally had in my right eye, along with a shift toward red in my left eye, immediately after the surgery on the right eye. It appears that my visual cortex went from being tuned for the previously impaired view in both eyes to the temporary mixed messages and on to the fully stabilized state with both eyes fixed.
    Great! Amazing software.

    I stand by by inference that the Sun, moved out to a few lightyears and viewed from above the atmosphere, would look neutral white and thus would be a suitable standard for defining a neutral white emitter.
    Yes, and I suspect most of the A, F and G stars are white, contrary to color use on the H-R diagram. Specifically when the A stars appear to gain a tint of blue and the G stars gain some yellow would be an interesting study. If this varies for stars off the main sequence would be interesting to learn as well. Both Vega and Sirius are A0 stars and both are main sequence stars (per Wiki), so perhaps this is close to the edge of gaining a blue tint.

    I should have noted above any reference I give to the Sun's color is based on it as an extended object. A star's "true color", IMO, should be a little like magnitudes: color as seen of the star as an extended object (ie disk) for "absolute" true color; point-source for "apparent" color. Point-sources will tend to favor a slight redshift likely due to the fact the fovea lacks blue color cones, though I don't recall reading anywhere that this has been properly tested.
    Last edited by George; 2018-Oct-29 at 02:42 PM.
    We know time flies, we just can't see its wings.

  27. #27
    Join Date
    Sep 2003
    Posts
    12,396
    BTW, I checked the RGB value for my avatar: 254, 254, 254. Can anything be whiter than that?

    Close to the limb the values are about 244, 237, 226 (desaturated orange), yet the limb region varies for some reason especially near the green plastic piece where it becomes are more destaturated green, so I suspect some glare from this piece may contribute to the greening effect near it.

    The other interesting fact is how flat the photon flux distribution is for the Sun's spectrum after converting it from its Planck-like SED. It's peak for the SED is in the middle of the blue region (contrary to many articles that say otherwise because they err by using a Planck distribution for the Sun and not a real sp. irr. distribution that are easily obtainable). So, it's a little ironic that when we convert the SED to a photon flux density distribution, the high blue region becomes the lowest level of flux because the blue photons carry more energy than the red ones so fewer photons exist in the blue region though their energy distribution is so much higher.
    We know time flies, we just can't see its wings.

  28. #28
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    Quote Originally Posted by George View Post
    I agree but we can demonstrate that it is less subjective than most people, including scientists, think.

    Galileo did not prove the Copernican model when his telescope revealed that Venus has both crescent and gibbous phases, but it did the one thing every proper scientific hypothesis and theory is vulnerable to -- falsification. He falsified the Aristotle/Thomist/Ptolemy theory.
    Irrelevant. This is objective geometric analysis, unrelated to how the human eye/brain system perceives pastel colors.
    It is a valid hypothesis to claim the true color of the Sun (as we would see it with neutral attenuation from space) is yellowish-white, or any other tint from white. After multiple efforts to disprove this, falsifying a yellow claim was something so simple that I all but lost interest in the entire topic. My avatar is, like Galileo's dual phases of Venus evidence, debunks any yellow tint claim for the Sun. It is an unfiltered projection of the Sun from the projection room of the world's largest solar telescope.
    I disagree. That avatar does not prove anything because it depends on the accuracy of the devices used to record and subsequently display the image, and in any case it does not take into account the atmospheric effects which at that altitude will be only about 25% less than at sea level.

    It's funny how I spent the entire night at KP (Kitt Peak)-- not to mention it was our 25th anniversary night (thanks Hun!) -- on a KP 16" RC scope trying to get solar twins like 18 Sco imaged with special techniques to tickle-out its true color, then to get a tour of the McMath-Pierce solar observatory without understanding what I had just observed. Perhaps I became more awake as we drove off when, all of sudden, it hit me like a ton of bricks when I realized that their white image totally destroys any claim for a yellow or yellowish Sun.
    You saw a presumably white projection surface illuminated by the Sun in a room that had other lights which you have mentioned in other posts. Too many variables not controlled.

    Dr. Potter and Roy Lorenz took this image for me, adding the RGB color pieces they scavenged-up to accommodate me to demonstrate it was a color image. I later bought and sent them a photometer to measure its intensity to demonstrate it is well within the normal photopic vision range. [Hornblower wisely questioned this earlier and I meant to mention this. Objects overly bright will disrupt true color determination.]

    There were many who questioned Galileo's wimpy telescope and the bold claims against a 2000 year theory, which got fused into their theology since Thomas Aquinas' work. His observations were not confirmed by all and there must have been good reasons why he saw things falsely. [The Jesuits, apparently, used a telescope and were fairly quick to confirm Galileo.]
    Once again, irrelevant to inquiries about color perception.

    So, if you, like most, take the Sun's color as something more of a subjective-type claim and not one that important to scientific progress -- it isn't really that important, admittedly -- then you are missing a moment to appreciate the scientific method at work. Please look at my avatar, think about the circumstances, and see how it destroys a yellowish claim. Ambient lighting must be considered, admittedly, but, like a fuzzy crescent Venus, such issues are easily resolved, IMO.

    Yes, though a black background should be the standard when stating a star's true color. But you will find that images of daytime solar projections will have a white Sun surrounded by a blue tint background due to the fact the the sky is blue and that is the color being reflected in the shadow of the telescope doing the projection. This blue background will tend to make some objects appear more yellow than normal due to the way the eye behaves, further arguing for a white Sun.
    That is not how I did my test projections last winter. I was in a blacked-out room with a 1-inch hole to admit the Sun's light to a 2-inch binocular objective right against the wall to keep any stray light from bypassing the binoculars and putting unwanted light on the screen. The only sky light to reach the screen was through the optical system and was certainly less than 1/100,000 as bright as the solar disk. With eyepiece projection I made the disk about 10 inches in diameter, which means the light was about 1/100 of direct sunlight and comparable to bright indoor lighting. My impression around noon was of pretty much neutral white with a slight yellowish or brownish cast toward the limb. That limb yellowing was confirmed quantitatively in my photo editing software. At this point I was more interested in the limb darkening and yellowing along with the variation of the color with elevation than with the absolute values
    [Example below from Purdue Engr.]

    PurvesCube.jpeg
    I don't see the relevance of these images.
    A contrary argument could be that a blue background in solar projection images (just Google them and there are probably hundreds) is more blue than normal because the camera color processing would take the brightest source and shift it, assuming it had a hint of yellow, to it being a white source, thus making it white. Cameras do this for a reason: that's what our brains do, so they are trying to represent what we see as color, and what we see is the definition of color; all camera-oriented color computer models strive to replicate what most of us already have.
    Once again, in any photograph that does not overexpose the solar disk and properly excludes stray light, the sky in the image will be black.

    If we can anchor the Sun as a white star, then it should help to open the minds of those who are misusing (unwittingly no doubt) the secondary color scheme found in the H-R diagram.

    I'm guessing the history behind this goes back to prominent leaders like Secchi.

    That's an important question and the answer is he did not.

    Le Soleil (The Sun) was perhaps the most significant book by Secchi. Unfortunately, I cannot find an English translation of this book but I did manage to translate his table of contents and found, in chapter 7, that he addressed his star types, which grouped primarily by spectral lines and secondarily by color (so I think this is the case. Anyone hear read French?) He admits that what we really see for Vega includes a hint of blue though he put it with white stars, though this may have been when he just had three types. He later included blue-white stars and I'm just being a little too rushed to dig-up the info on it right now.
    OK, he used simplified color names for classification purposes and others have taken it too literally. I would not fault him for that.

    Yep, I think that's hitting the nail on the head. Color gives the 30,000 ft. view of things when we are now in a world of 10' observations that interest us. Once the importance of the spectral lines became understood AND, almost simultaneously, black and white spectrographs ruled the day, color became just another enjoyable and artistic touch to things. Nevertheless, S&T should have stayed at least within the 30,000 ft. range and not jump to over 300,000 ft. without telling anyone their altitude.
    I don't know how you got that line of thought. Jumping to the edge of outer space will just shift the hues a bit toward the blue and leave them very pale. What the magazine did was shift the hues toward the red and oversaturate those from G through M, along with oversimplifying them.

  29. #29
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,265
    Quote Originally Posted by George View Post
    That's interesting and it's nice to see the account of this given how the blue background could be argued to have a yellowing effect on the star. There are arguments as such as to why the Sun often has a tint of yellow during a very quick glancing view of it high in the sky. I'm guessing the power of color constancy is more remarkable than we guess.

    Nice report. I notice you have a tint of blue for Vega but not for Sirius though they have the same B-V value. I will guess that it's due to the large altitude difference during the daytime; Vega being very close to zenith, I think, at one point. Is this likely or do you think the daytime vs. nighttime viewing is the better account?
    What is all of this about daytime viewing of the stars? My new observations were last night around 10 PM, with all of the stars well up in the sky. The ones in post 24 were in March 2015, with Orion and Sirius near culmination. I would have had to wait until the wee hours to re-observe them last night. I would expect the contrast with the blue sky in daylight viewing with a telescope to cause an apparent shift toward red, just as it does with Betelgeuse in twilight.

    Great! Amazing software.

    Yes, and I suspect most of the A, F and G stars are white, contrary to color use on the H-R diagram. Specifically when the A stars appear to gain a tint of blue and the G stars gain some yellow would be an interesting study. If this varies for stars off the main sequence would be interesting to learn as well. Both Vega and Sirius are A0 stars and both are main sequence stars (per Wiki), so perhaps this is close to the edge of gaining a blue tint.

    I should have noted above any reference I give to the Sun's color is based on it as an extended object. A star's "true color", IMO, should be a little like magnitudes: color as seen of the star as an extended object (ie disk) for "absolute" true color; point-source for "apparent" color. Point-sources will tend to favor a slight redshift likely due to the fact the fovea lacks blue color cones, though I don't recall reading anywhere that this has been properly tested.

  30. #30
    Join Date
    Sep 2003
    Posts
    12,396
    Quote Originally Posted by Hornblower View Post
    Irrelevant. This is objective geometric analysis, unrelated to how the human eye/brain system perceives pastel colors.
    There were many who would not look through Galileo's telescope for fear that they would be fooled by what they saw. It wasn't that great a telescope, after all. Of course, his telescopes did improve and higher magnifications came along as well.

    So I believe my avatar claim is analogous.

    I disagree. That avatar does not prove anything...
    Nor did I claim it was "proof". Hypotheses are not about proving but their falsification is the backbone of science. I claim that my avatar falsifies a yellow Sun claim. I am hoping readers will consider what I am saying and reason themselves why my claim is likely sound. It shouldn't be that hard, though I am open to criticism on where I could be wrong.

    ...because it depends on the accuracy of the devices used to record and subsequently display the image,
    What would be the level of accuracy needed? As to what was seen, the picture is a very accurate one as to what a "normal" vision would behold in the projection room; WYSIWYG. The RGB plastic pieces also serve as color correcting references. These serve to adjust for any ambient lighting problems, if we were to put more effort in doing so, admittedly. I think your questioning whether or not the ambient lighting is fooling both eye and camera is a fair question. The abundance of white disks from other methods discounts this possibility though the KP image is so elegant compared to any other, a focal ratio of 50 is quite nice. [I noted earlier that the intensity was checked, so we can eliminate photopic level issues.]

    and in any case it does not take into account the atmospheric effects which at that altitude will be only about 25% less than at sea level.
    [my bold] Exactly! That is the crux to the yellow falsification. Do you see why? It took me a while to see it, yet it's so simple.

    You saw a presumably white projection surface illuminated by the Sun in a room that had other lights which you have mentioned in other posts. Too many variables not controlled.
    Again, that's a fair response if it weren't for the fact that it isn't the only solar projection that exists. Secondly, again, the color pieces are useful color references to adjust if needed for ambient lighting.

    A simple pinhole projection is nice since it avoids lens chromatic issues, though diffraction can be an issue and I suspect can produce more yellowish limbs than normal. A pinhole size of about 1.2mm -- the rounder the better -- will resolve at 1 meter and is close to optimal as a trade-off between diffraction and resolution. Then there are projections from telescopes. I like the Sunspotter image as it is easy to set-up and makes for a larger than normal disk size, compared to other pin-hole designs. Here's one I took...

    Reduced size image from Sunspotter.jpg

    Under certain circumstances some disks will have a yellow limb, but with normal vision on a clear day with clean air and a high Sun, then I will bet an ice cream Sundae it will be a white (non-yellow tint) disk every time, lens chromatic issues not withstanding though a single lens, as in the Sunspotter, shouldn't be an issue.

    Once again, irrelevant to inquiries about color perception.That is not how I did my test projections last winter. I was in a blacked-out room with a 1-inch hole to admit the Sun's light to a 2-inch binocular objective right against the wall to keep any stray light from bypassing the binoculars and putting unwanted light on the screen.
    Yes, and you did mention this earlier; sorry I missed it. Were you using a single objective or passing it through all the lenses in the binocular?

    The only sky light to reach the screen was through the optical system and was certainly less than 1/100,000 as bright as the solar disk. With eyepiece projection I made the disk about 10 inches in diameter, which means the light was about 1/100 of direct sunlight and comparable to bright indoor lighting. My impression around noon was of pretty much neutral white with a slight yellowish or brownish cast toward the limb. That limb yellowing was confirmed quantitatively in my photo editing software. At this point I was more interested in the limb darkening and yellowing along with the variation of the color with elevation than with the absolute values.
    A 10" projection is very nice, especially if sunspots ever return. Are you confident that the yellowish limb is not from lens chromatic issues or diffraction? It's certainly possible that it is an accurate projection, though all the particulate matter can also bump the AM value. The golden eta Cas. was an eye-opener for me.

    I don't see the relevance of these images.
    Since solar disk projections include a blue background and ambient blue sky light, how this effects one's color determination is possibly relevant. Note the differences in the colors of the squares as shown. A blue background will tend to make us see more yellow than is actually there. Yet even this effect does not alter a white disk result. In your dark room case, you've eliminated this issue, fortunately. [I tried to make a similar room using black plastic around a canopy and learned duck tape will not work on black plastic in 105F temperatures.]

    OK, he used simplified color names for classification purposes and others have taken it too literally. I would not fault him for that.
    I also recall reading somewhere that the "Pickering Computers" suggest that the star colors might correlate with temperatures and spectral lines. It seemed to be presented as something curious and not that important to them, and I don't recall if it was Murray or Cannon or just who it was that discussed it. Once the spectral lines were appreciated, the colors apparently became more art than science, causing their suggestion of being scientific to get abused. So the H-R diagram coloring gets abused and I doubt this was ever their intention, but I am unsure of this.

    I don't know how you got that line of thought. Jumping to the edge of outer space will just shift the hues a bit toward the blue and leave them very pale. What the magazine did was shift the hues toward the red and oversaturate those from G through M, along with oversimplifying them.
    It was just a metaphor of a metaphor used for poor resolution (inaccurate over-saturation). The "30,000 ft." metaphor is popular these days.
    Last edited by George; 2018-Oct-29 at 06:43 PM.
    We know time flies, we just can't see its wings.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •