Michael A. Covington    Michael A. Covington, Ph.D.
Books by Michael Covington
 
Previous months
About this notebook
Search site or Web
Ichthys

Daily Notebook

Popular topics on this page:
How long should astrophoto exposures be?
Astrophotos:
Moon
M13 with dust lanes
Many more...

This web site is protected by copyright law. Reusing pictures or text requires permission from the author.
For more topics, scroll down, press Ctrl-F to search the page, or check previous months.
For the latest edition of this page at any time, create a link to "www.covingtoninnovations.com/michael/blog"

Ads by Google, based on your browsing history
 
 
2020
June
6

We must oppose racism, not just avoid it

Many of us have had our attention drawn to the way race relations in this country have, in many ways, backslid or at least stagnated since the 1970s. Our liberal wing lost interest in race when it stopped being controversial, and our conservative wing became alarmingly tolerant of reactionary racism.

We are realizing that it is not enough to avoid practicing or promoting racism; we must actively oppose and counteract it.

When we say "Black lives matter" we do not mean other lives do not matter. We mean there is a specific hazard to black lives that needs to be fixed.

I have two particular concerns. One is no longer letting white supremacists pass themselves off as political conservatives. We need to tighten up the definition of conservative so that it doesn't include them.

The other is doing away with Confederate nostalgia. Many of us were miseducated, in elementary school in the 1960s, to believe that the Confederate flag stood (only) for regional pride and self-government. But in fact the Confederacy was formed to preserve slavery. It is not a Noble Lost Cause. It was a terrible mistake, although of course once it got going, many individuals within it meant well and did noble things. That does not legitimize the cause itself, any more than Russian heroism in WWII legitimizes Soviet communism.

"But General Lee!" I respond: We can honor the professional achievements and, later in their lives, the good U.S. citizenship of both Robert E. Lee and Wernher von Braun without flying the flags under which they fought in the war. Confederate flags are not welcome where I go.

2020
June
4

The videoed riots

The riots of recent days, and the peaceful demonstrations that preceded them, and even the crimes that provoked the demonstrations, are like none other in the history of the world, for a reason that has not yet been fully appreciated.

They were photographed and video-recorded constantly, by people on all sides of all controversies, and the images have been preserved.

For months, we will be finding out more about what really happened.

I think digital image sensors have been a powerful force for good in recent history. From police body cameras to civilian smartphones, they have made it possible to preserve whatever the eye can see. If the police cameras happen to all be turned off, civilian cameras are still on.

There is more evidence than anyone ever wished or feared.



Farewell, world's last working CDC 6000-series

The first computer I ever programmed was a Control Data CDC 6400, back in 1973. Thirty-one years later, in 2004, I was amazed to find something similar (a later-model CYBER 960) still running in Germany, and I was able to try it out remotely (click here).

The other day I got word that a CDC 6500 (like the 6400 but with two CPUs) was still running at the Living Computers Museum in Seattle and could be accessed through a terminal program that runs in a web browser.

I was all set to take it for a spin when tragic news reached me: The Living Computers Museum is shutting down, perhaps permanently.

Its CDC 6500 page is archived on The Wayback Machine here.

[Added:] What's worse, digging deeper, I find that the Museum announced in March that their 6500 had broken down with a cooling system failure, and as far as I know it has not been repaired.

It appears, however, that the CYBER 960 that I used in 2004 may still be operating; web page here.

A CDC 6400, or possibly its successor the CYBER 70/74, was involved in the day Melody and I met.



Novis oculis

Well, at least novo oculo (singular). We are thankful that Melody's cataract surgery on Monday went very well. Her left eye had developed a cataract that disorted the shape of the lens and introduced a focusing error that was not correctable even with glasses. No more! She is still going to need glasses and is waiting for the new lens to settle down so she can get them; in the meantime I am the chauffeur, but she's enjoying the view!

2020
June
3

Troubled times

Those of us who oppose racism should do it in a thousand or a million different ways, to soak into all the cracks and pores of a society that needs reforming. We should not be expected to accept a single political platform, sociological theory, wordy statement, slogan, or one-size-fits-all solution. We should each do the good we can, where we are, among the people with whom we have contact, and no two situations are alike. What unifies us is respect for our fellow human beings. What matters is not how much power we have, but whether we are always exerting our influence in the right direction.

2020
June
1

Moon, unexpectedly sharp

Picture

This was just a casual snapshot taken while setting up the telescope, but it came out well. The quarter moon was high overhead as I took aim at it with my Celestron 8 EdgeHD and f/7 reducer. Because this optical configuration has a wide, flat field, it gave me a picture of the whole face of the moon that is sharp all over, often hard to achieve with conventional astronomical telescopes, which are sharpest in the center.

The other thing that surprised me is that even though it doesn't have electronic first curtain, the Nikon D5500 had surprisingly little shutter vibration, in fact none noticeable here. The fact that the exposure was short, 1/500 second, helped with that.

This is a color picture, and the saturation hasn't been manipulated; the color is realistic. If you look closely, you can see that some of the gray areas (maria) are more brownish and others are more bluish. This is due to the presence of different minerals, and once it is pointed out, the color difference is easy to see in any telescope or even binoculars, but visual observers ignored or denied it until Apollo started bringing back samples!



M13 with dust lanes
Using PHD2 to autoguide with a CGEM

Here is the great globular cluster M13 in Hercules, with visible dust lanes (streaks of interstellar dust) in front of it. The second picture shows where to look for them. The upside-down-Y-shaped structure at the left is popularly known as "the propeller."

Picture

Picture

Stack of 18 2-minute exposures, Celestron 8 EdgeHD (200 mm aperture), with f/7 focal reducer (1400 mm focal length), H-alpha-modified Nikon D5500 at ISO 400.

This picture reflects long-awaited success in getting PHD2 guiding software to cooperate with my CGEM mount. Celestron mounts are designed to have a lot of backlash, and after carefully adjusting the CGEM, I was able to get its declination backlash down to a mere 4000 milliseconds of guiding movement (that's how PHD2 measures it). Part of the secret is setting the declination guide rate to 90% of maximum (the RA rate should remain at 50%). Then turn on declination backlash compensation in PHD2 and let it adjust itself. Recent improvements to PHD2 have helped.



What works in practice also works in theory
My DSLR exposures have been the right length all along

I wish I could remember who referred me to Dr. Robin Glover's video about astrophotography with CMOS cameras so I could thank them. (If they pop up and claim credit, I'll gladly edit this entry.)

The video mostly addresses the question of how long to expose, and the take-home lesson is that if your sky background is showing up moderately bright, then you're exposing long enough, and thermal noise and read noise are unimportant because they are swamped by the shot noise of the sky background (which would be there with any camera).

So it really is as good to stack ten 2-minute exposures as to make one 20-minute exposure. Preferable, in fact, because a single long exposure risks guiding problems, interruptions by aircraft, and overexposure. For the best dynamic range, you want a large stack of the shortest exposures that get the sky background up to a reasonable level. And that level depends on your camera.

Dr. Glover goes through the calculations, using mathematical models of noise and of the sky background. Using his formula and assuming the read noise in electrons is 2.5 (which is typical for a Nikon D5500 at ISO 400 or a Canon 60D or 80D at ISO 1600), here are the recommended exposures:

Rural site
(Bortle 3)
My house
(Bortle 7)
f/4 3 min 20 sec
f/7 7 min 2 min

Expose 1/3 as long if you are using a monochrome sensor, because every pixel gets the whole spectrum, not just a third of it. Expose much longer if you are using a narrowband filter.

These are ballpark figures; half or twice as much exposure is practically as good. The point is, the calculations show us that we don't need to go for 20 minutes, and, conversely, that 5 seconds would be too little.

The reason the exposure recommendations are not exact is that judgments can differ as to how much noise is noticeable in a picture. (Remember that the issue here is not whether there's noise, but rather whether you'd get visibly less of it with longer exposures.)

To get DSLR sensor noise levels in electrons, use the chart at PhotonsToPhotos. With a noisier sensor, you need to expose to get the sky background brighter. That is also the case if it's warmer and there is more thermal noise.

All of this accords very well with how we've been doing it all along: Expose so the sky background is moderately bright, definitely distinct from black. With a noisier sensor, you need to expose for a brighter sky background, probably using a higher ISO setting. Here's my classic example of how the histogram on the camera should look:

Picture

Crucially, the hump isn't touching the left edge. It can be anywhere from the position shown, all the way over to the middle; it should not be farther to the left unless you have an unusually low-noise sensor.

The biggest surprise that emerged from this, for me, is that at the settings needed for best performance, there's not much difference between second-generation and third-generation Canon DSLRs. Sony (including Nikon) still has a considerable edge. And of course if you can work at ISO 400 rather than 1600, you have four times the dynamic range.

Another of Dr. Glover's insights is that there is less need to cool your sensor than you probably think. If you are having trouble getting enough light to the sensor (because of a narrowband filter or a high f-ratio), or if you have an unusually dark sky background (think Arizona desert or Australian Outback), then expose longer and use a cooled sensor. But if you're like most of us, you might be better off not bothering with it. Newer sensors have much less thermal noise than older ones whether CMOS or CCD (see around 18:35 in the video). I've heard from people testing late-model DSLRs who were quite surprised at how little the performance depended on temperature.

And the biggest takeaway is that we're usually not working near the limits of our sensors. If a lot of different cameras all seem to be working about equally well, it's because they really are. By experimenting, we've already more or less found out how to get the best out of each of them.


If what you are looking for is not here, please look at previous months .