126 replies to this topic

[Smitty14]

Greetings all,

 

I was fortunate to be in Nebraska for the totality for the 2017 eclipse.  Since that was my first time being in totality, I just wanted to view the 2 minutes and 36 seconds without worrying about my cameras and technology.

 

So, this April I do plan to image the eclipse using Solar Eclipse Maestro. I purchased an older Mac on ebay for $75.00 and loaded the software. 

 

I have found a few scripts online from some generous people who have posted their scripts.  Attached is the one that I tested and am happy that the camera shutter responded to the simulation.

 

I would very much appreciate if other people had success, or have improved upon their scripts that were used in previous eclipses and share their work.

 

Clear Skies!

 

Mark Smith

 

 

 

 Final R5 Script.txt   7.46KB   329 downloads

[starrycanuck]

Greetings all,

 

I was fortunate to be in Nebraska for the totality for the 2017 eclipse.  Since that was my first time being in totality, I just wanted to view the 2 minutes and 36 seconds without worrying about my cameras and technology.

 

So, this April I do plan to image the eclipse using Solar Eclipse Maestro. I purchased an older Mac on ebay for $75.00 and loaded the software. 

 

I have found a few scripts online from some generous people who have posted their scripts.  Attached is the one that I tested and am happy that the camera shutter responded to the simulation.

 

I would very much appreciate if other people had success, or have improved upon their scripts that were used in previous eclipses and share their work.

 

Clear Skies!

 

Mark Smith

 

 

 

Final R5 Script.txt

Thanks for passing this on. Interesting, it looks exactly like an Eclipse Orchestrator script … do (did) Maestro and Orchestrator share a common code base?

[Smitty14]

starrycanuck,

 

Good question.  I am not familiar with the script for Orchestrator, and will leave that for someone much more knowledgeable than I! 

 

Mark

[mr.stratosphere]

Hi Smitty,

 

For the 2017 total eclipse, I used the script from Jerry Lodriguss on a vintage MacBook Pro.

 

See it here:  https://www.astropix...-safety-script/

 

I strongly encourage you to carefully look through all of his eclipse pages for valuable ideas and suggestions.

 

Good Luck!

 

Jim in Oregon / mr.stratosphere

[Smitty14]

Hi Smitty,

 

For the 2017 total eclipse, I used the script from Jerry Lodriguss on a vintage MacBook Pro.

 

See it here:  https://www.astropix...-safety-script/

 

I strongly encourage you to carefully look through all of his eclipse pages for valuable ideas and suggestions.

 

Good Luck!

 

Jim in Oregon / mr.stratosphere

Jim,

 

Thank you so much!  I will check it out.

 

Mark

[szigler511]

Hi Mark,

 

Thanks for sharing your SEM script! I used SEM in 2017 and in 2019. Each time I employed two Nikon DSLRs, one attached to a 100 mm refractor and the other on a telephoto zoom lens. Below I'm sharing my script from 2019 when I used two D850s. The D850 on the telescope is referred to as "D850-SC" and the camera on the telephoto lens is "D850-02". I tested the script extensively beforehand with an XQD card in each camera (no SD card) and it worked perfectly on eclipse day. Unfortunately, I made the mistake of leaving time gaps at critical moments just before C2 and just after C3. These gaps resulted in a discontinuous series for the Baily's beads composite I had envisioned. I’m currently working on a better option for April but it is still a work in progress.

 

Steve

 

[szigler511]

 Eclipse Day 2019 V7-TwoD850_CN.txt   8.27KB   158 downloads

[DAA]

With a vintage Mac running Mojave (still alive despite replacement by a M1 MacBook Pro) and with a Nikon D850 (with 400mm + 1.4TC), I really appreciate this thread.  I'll be reading the scripts from experienced automated eclipse shooters (specifically for SEM) and experimenting with the sun to learn and to test reliability/reproducibility.  I'll use an auxiliary battery with the vintage Mac.  Collaboration on SEM scripts, started above, makes a very good start for new SEM users like me.  I look forward to commenting more in this thread in the near future.  All this to level up from straight manual shooting that I did for the 2017 TSE which, while adequate, melted down when I gazed up into totality.  In the end, I'd like to create a wide field HDR image from Earthshine and > 8 solar diameters to corona.  Thanks to all contributors above!

[Smitty14]

Hi Mark,

 

Thanks for sharing your SEM script! I used SEM in 2017 and in 2019. Each time I employed two Nikon DSLRs, one attached to a 100 mm refractor and the other on a telephoto zoom lens. Below I'm sharing my script from 2019 when I used two D850s. The D850 on the telescope is referred to as "D850-SC" and the camera on the telephoto lens is "D850-02". I tested the script extensively beforehand with an XQD card in each camera (no SD card) and it worked perfectly on eclipse day. Unfortunately, I made the mistake of leaving time gaps at critical moments just before C2 and just after C3. These gaps resulted in a discontinuous series for the Baily's beads composite I had envisioned. I’m currently working on a better option for April but it is still a work in progress.

 

Steve

Steve,

 

I would love to see the updated script that fills in the gaps.  Thank you for sharing your work!

 

Mark

[Goldenberg]

My Nikon Z6 camera when commanded from a script in Solar Eclipse Maestro ( will skip frames and slow down and sometimes not update the exposure per the script if it is commanded at more than a photo every 1.5 sec or so.  The camera itself is capable of 10 (or maybe 12 frames per sec - I forgot). I want to speed up the camera to get lots of shots of the diamond ring and Bailey beads.  I know the buffer on the Z6 is only about 45 frames or so but that is enough. I would also bracket some exposures if I could shoot faster. 

Is there anything I can do to reliably get more shots per sec?

Edited by Goldenberg, 22 February 2024 - 09:32 PM.

[MikeCMP]

My Nikon Z6 camera when commanded from a script in Solar Eclipse Maestro ( will skip frames and slow down and sometimes not update the exposure per the script if it is commanded at more than a photo every 1.5 sec or so.  The camera itself is capable of 10 (or maybe 12 frames per sec - I forgot). I want to speed up the camera to get lots of shots of the diamond ring and Bailey beads.  I know the buffer on the Z6 is only about 45 frames or so but that is enough. I would also bracket some exposures if I could shoot faster. 

Is there anything I can do to reliably get more shots per sec?

Check your scripting commands, if they are using Mirror Lock up (MLU) this will limit the max speed of the camera by a decent amount. I know its to limit "vibration" but for gods sake we are talking about shots well under fractions of a second in most cases, vibration can't b that big a deal.

 

I am planning to mount my camera on a EQ-6 though so that will probably help reduce the vibrations as well. 

 

If you go through the configuration wizard, the script it creates will not have MLU, then if you load it and do a simulation from your location you plan to image from and select simulate from second contact you can see how fast it will take the shots.

 

I am working through this video at the moment i recommend it 

 

https://www.youtube....h?v=qLHfURDO1YY

 

Mike

[Goldenberg]

Mike

 

Thank you for your post.  My camera is mirrorless so mirror lockup should not be a problem.  In my script for MLU I put in 0.0 which means no mirror lockup.  I think I have that covered

 

I have discovered my camera has a mechanical shutter and an electronic front. curtain shutter.  The skipping and slow down seems to occur only in electronic front curtain shutter.  The problem here is that sometimes the camera switches on its own from mechanical shutter to electronic.  The Nikon Z6 camera mechanical shutter is capable of speed as fast as 1/8000 but when commanded by script I can only get the camera to go as fast as 1/2000.  For some reason sample scripts use 1/8000 for prominences, chromosphere, and Bailey Bds.  Not sure why they need such a high shutter speed but I will be using 1/1600 at f13.0 because of my camera limitation.

 

I am investigating the camera problems. 

 

David

Edited by Goldenberg, 24 February 2024 - 03:11 PM.

[Goldenberg]

In his presentation on YouTube (https://www.youtube....h?v=qLHfURDO1YY)    Xavier mentions the camera will fire more reliably with a serial port connection to trip the shutter. Does the present version of the software support using a serial port connection for the shutter?  I guess the USB connection would still transmit the exposure information. 

[Smitty14]

My R5 Script above is working very well in simulation.  However, I would like to add bracketed images 3 minutes apart during C1-C2 and C3-C4.  Can anyone help me with this?

 

Thank you so much.

 

Mark

[agavephoto]

What about setting up three INTERVALOMETER loops slightly offset from each other? I chose 25 frames to leave a little time for prep for diamond ring and all that fun just before C2. Then, just have the opposite starting however long after C3 you want to start the C3-C4 sequence.

 

#Bracketing 3 exposures every 3min, 1s apart, starting 3 minutes before C1

#

FOR,(INTERVALOMETER),0,180,25

TAKEPIC,C1-,00:03:00.0,R5, exp1

ENDFOR

#

FOR,(INTERVALOMETER),0,180,25

TAKEPIC,C1-,00:03:01.0,R5, exp2

ENDFOR

#

FOR,(INTERVALOMETER),0,180,25

TAKEPIC,C1-,00:03:02.0,R5, exp3

ENDFOR

 

 

My R5 Script above is working very well in simulation.  However, I would like to add bracketed images 3 minutes apart during C1-C2 and C3-C4.  Can anyone help me with this?

 

Thank you so much.

 

Mark

 

 

[agavephoto]

Has anyone found the right incantation to get SEM to put the Canon R5 into high speed shutter mode? The SEM docs suggest setting 3 for slow on Canon, 100 for high speed mode. Using 100 works fine for the 5D4, but I can't seem to get the R5 into high speed mode via SEM.

[DAA]

I have been staring at SEM's deluxe.txt that is described as an aggressive script.  It certainly is, cramming in about 14 images in the 8s surrounding C2 and repeating this in a short time interval surrounding C3.  To accomplish this, some of the times between images (including camera parameter changes) are as short as 0.4s and all of the times between images in these intense sections of the script are less than 1s.  From what I see with my Nikon D850 with an XQD card, that poses some challenges.  I'm looking for advice.

 

First, using mirror lock-up of 0.2s or even 0.0s, my camera cannot keep up.  It tries, but ends up dropping some frames.  If I execute a mirror lock-up command LVPSTART before the rapid sequence, then SEM crashes (!) after a couple of frames.  It does this reliably -- no getting around it.  I can capture a single frame between an LVPSTART and LVPSTOP command, but if I try to stuff several frames between, I cause a crash.

 

I now know that a CFExpress type B card has considerably faster write speeds than my circa 2018 XQD card.  But is writing to memory (or overwriting memory) causing the crash?  Or is the time to reset camera parameters such as ISO / F-number / exposure time causing the issue?  If the problem is write speed, then maybe a faster card would help.  (Certainly, that's what Xavier suggests by recommending an 'Extreme IV memory card' which today is slow by comparison to CFExpress type B cards.)  But if it is time to reset camera parameters, I'm not sure how to realize such a dynamic imaging sequence as shown in deluxe.txt.

 

One compromise solution is just to use burst mode and not change camera settings.  Another compromise solution is to back off and leave more than 1s between images, sacrificing some of the HDR potential of switching camera settings.

Has anyone successfully achieved the frame rates (including camera changes) required by Xavier's deluxe.txt script?  I'm assuming that a serial cable is not required based on previous posts.  I'd really like to run for a few hot seconds with mirror lock-up so that mirror slap in my D850 doesn't smear the few stars that might appear in the frame but given the crashes that I'm causing, no way at this point.  Buy a new card?  Back off?

 

Please advise.
Thank you,
Dave

[DAA]

Quick follow-up on SEM crashes with rapid frame rates (and MLU).  

 

I said that the crashes occurred reliably but with the following pseudo-script, I refined the range when crashes occurred 'some' or 'most' of the time.

 

LVPSTART

take 4 images with changing ISO / F-number / exposure time spaced by X seconds

LVPSTOP

 

With X = 1s, 0.9s, 0.8s, 0.7s, there were no crashes.

With X = 0.6s, crashes occurred once in a while.

With X = 0.5s, crashes occurred half the time.

 

From which I conclude that pre-mirror lock up allows more rapid frame rate than using MLU with each frame.  And the faster frame rate gets into a trouble range with MY camera / card combination (Nikon D850 / XQD) when the frame rate exceeds 0.6s / frame.  I still don't know if the crashes are related to exceeding possible rate of image data to memory buffer or related to changing camera parameters.  So, I'm inclined to look into a new card with the hope that this solves the problem.  (Note, Xavier says that his deluxe.txt script works even with X = 0.4s and without MLU!)

 

Back to practicing.

Edited by DAA, 09 March 2024 - 02:37 PM.

[DAA]

Hi agavephoto -- that is a clever way of bracketing periodically through C1 to C2.  Using your method, exp1, exp2 and exp3 can be whatever you want rather than the slightly more constrained values that you'd get with TAKEBKT, BKTDATA and TAKEBST where the three exposures would be separated by a given number of EV.  I like it.  

 

I guess you could use MAGPRE and MAGPOST as well, but why if you've already figured out how many and at what interval you want your frames?

[agavephoto]

Has anyone found the right incantation to get SEM to put the Canon R5 into high speed shutter mode? The SEM docs suggest setting 3 for slow on Canon, 100 for high speed mode. Using 100 works fine for the 5D4, but I can't seem to get the R5 into high speed mode via SEM.

Update: looks like SEM cannot control the burst sequences properly for the R5 even at slow speed: it puts the camera in to an endless loop at one of the shutter speeds and won't stop until the camera is powered off if the script ends on a burst. Xavier said the beta version didn't fully support the R5, so I'm not shocked at the limitations.

 

The R5 goes through the RAMPUP sequences faster than the 5D4, so I can get more total exposures if I stick the ramps set on the R5. I'm trying to convince myself whether ramp sets repeated as many times as possible is better or worse than taking short exposures early on, longer exposures close to the max, and then more shorter as C3 approaches.

[Smitty14]

What about setting up three INTERVALOMETER loops slightly offset from each other? I chose 25 frames to leave a little time for prep for diamond ring and all that fun just before C2. Then, just have the opposite starting however long after C3 you want to start the C3-C4 sequence.

 

#Bracketing 3 exposures every 3min, 1s apart, starting 3 minutes before C1

#

FOR,(INTERVALOMETER),0,180,25

TAKEPIC,C1-,00:03:00.0,R5, exp1

ENDFOR

#

FOR,(INTERVALOMETER),0,180,25

TAKEPIC,C1-,00:03:01.0,R5, exp2

ENDFOR

#

FOR,(INTERVALOMETER),0,180,25

TAKEPIC,C1-,00:03:02.0,R5, exp3

ENDFOR

Thank you!  I will give it a try.

 

Mark

[whwang]

From which I conclude that pre-mirror lock up allows more rapid frame rate than using MLU with each frame.  And the faster frame rate gets into a trouble range with MY camera / card combination (Nikon D850 / XQD) when the frame rate exceeds 0.6s / frame.  I still don't know if the crashes are related to exceeding possible rate of image data to memory buffer or related to changing camera parameters.  So, I'm inclined to look into a new card with the hope that this solves the problem.  (Note, Xavier says that his deluxe.txt script works even with X = 0.4s and without MLU!)

 

Hi David,

 

Like you, I also plan to use D850.  I just did some stress test on it.  I managed to take 120 exposures (1/160 sec) in about 36 sec.  SEM did not crash.  Below is the test script that I use.

 

LVPSTART,C2,-,00:00:19.0,D850, , , , , , , ,Start photography in Lv
CHGCSSL,C2,-,00:00:17.0,D850, , ,4, , , , ,Set frame rate to 4 per second
# Max number of exposures allowed by TAKEBST seems to be 20.
TAKEBST,C2,-,00:00:15.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,-,00:00:09.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,-,00:00:03.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,+,00:00:03.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,+,00:00:09.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,+,00:00:15.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
LVPSTOP,C2,+,00:00:22.0,D850, , , , , , , ,Stop photography in Lv

 

Basically, I set frame rate to 4 frames per second.  I asked TAKEBST to take a burst of 20 exposures each time.  So the time span should be 5 second.  Then I give each TAKEBST a 1 second "break" before the next one is launched.  This got repeated six times and there was no crash.

 

I found the break between each TAKEBST is extremely important.  If SEM is asked to launch a new burst sequence before an old one is complete, it can crash.  I kind of worried that 1-second breaks were not enough.  But it seems fine.  Next I will see if I can push it to 5 frames per second, and if this requires a longer break.

 

For your reference, I am using a Sony XQD with 400 MB/s of write speed.  I shoot in 14 bit mode with lossless compression.  

 

Let's discuss more.  I find the whole thing very tricky.

 

Cheers,

Wei-Hao

[agavephoto]

Hi David,

 

Like you, I also plan to use D850.  I just did some stress test on it.  I managed to take 120 exposures (1/160 sec) in about 36 sec.  SEM did not crash.  Below is the test script that I use.

 

LVPSTART,C2,-,00:00:19.0,D850, , , , , , , ,Start photography in Lv
CHGCSSL,C2,-,00:00:17.0,D850, , ,4, , , , ,Set frame rate to 4 per second
# Max number of exposures allowed by TAKEBST seems to be 20.
TAKEBST,C2,-,00:00:15.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,-,00:00:09.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,-,00:00:03.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,+,00:00:03.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,+,00:00:09.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
TAKEBST,C2,+,00:00:15.0,D850,1/160,15,200,20,RAW,None,N,Solar corona Rs = 0.1
LVPSTOP,C2,+,00:00:22.0,D850, , , , , , , ,Stop photography in Lv

 

Basically, I set frame rate to 4 frames per second.  I asked TAKEBST to take a burst of 20 exposures each time.  So the time span should be 5 second.  Then I give each TAKEBST a 1 second "break" before the next one is launched.  This got repeated six times and there was no crash.

 

I found the break between each TAKEBST is extremely important.  If SEM is asked to launch a new burst sequence before an old one is complete, it can crash.  I kind of worried that 1-second breaks were not enough.  But it seems fine.  Next I will see if I can push it to 5 frames per second, and if this requires a longer break.

 

For your reference, I am using a Sony XQD with 400 MB/s of write speed.  I shoot in 14 bit mode with lossless compression.  

 

Let's discuss more.  I find the whole thing very tricky.

 

Cheers,

Wei-Hao

Your comment in your script about the max # of exposures for TAKEBST is curious. I have been using over 20 without significant issue so far with my Canon 5D4, so perhaps you can increase this if you can get the frame rate higher or give no command after as a test? If I set my burst to 30, I get 28 frames, so I am dropping some in my testing. 

Edited by agavephoto, 12 March 2024 - 09:48 AM.

[whwang]

Your comment in your script about the max # of exposures for TAKEBST is curious. I have been using over 20 without significant issue so far with my Canon 5D4, so perhaps you can increase this if you can get the frame rate higher or give no command after as a test? If I set my burst to 30, I get 28 frames, so I am dropping some in my testing. 

 

Indeed, it's strange.  I started with just one line of TAKEBST, and a relatively low frame rate, just to be save.  Then I found that I cannot go beyond 20 exposures even if I varied the frame rate.  That's why I put such a comment there.  Before this, I thought the limit would be 100 exposures.

 

Although I don't like such a limit, in practice, I probably won't take a burst for much more than 20 exposures anyway.  This is because I want to mix a few different exposure times during the totality for the inner and outer corona.  So I will give each burst a short interval (5 to 10 seconds) and then switch to another burst of a different exposure time.  If I do so, there isn't really a strong need for a burst series to go much farther beyond 20 exposures.

 

Anyway, I will keep trying.

[DAA]

Hi Wei-Hao,

 

Thank you for showing your stress test script.  I am using identical hardware, Nikon D850 and XQD card with 400MB/s write speed + 14-bit, uncompressed raw images.

 

A key difference, I think, is that I am changing parameters between each shot, e.g., ISO, F-number and exposure time.  Here are two observations.  First, crashes to not occur if I do not use live-view to raise the mirror before the rapid sequence but in this case, frame rate is limited by the mechanics of the mirror.  I time the sequence by photographing a clock (https://clock.zone) with milliseconds using short exposures to not blur the numbers.  Second, crashes DO occur if I start the sequence by raising the mirror using LVPSTART (and ending with LVPSTOP).  I run into problems by shortening the interval between shots to about 0.6s or less.

 

Here are two failed attempts to fix the problem.

1.  Turn off all extraneous camera processes like 3D face tracking, matrix metering face detection, etc.  I might not have found all such processes but with all the ones I turned off, I still experience crashes.

2.  Upgrade my XQD card to CPExpress type B card with minimum write speed of 2400MB/s.  Still experience same level of crashing.  I will return the new card.  I might not return the card.  ;-)

 

I will try your stress test and hope/expect to get your results suggesting that our hardware behaves the same.  Below is the script that crashes SEM about half the time.  I would be interested if you have the same problem.  This script is not very different from the hectic part of Xavier's deluxe.txt, I think.

 

LVPSTART,C2,-,00:22.0,D850, , , , , , , ,start in live view, mirror up
TAKEPIC,C2,-,00:20.0,D850,1/1000,4.0,6400,0.0,RAW,None,N,#1
TAKEPIC,C2,-,00:19.4,D850,1/800,5.6,6400,0.0,RAW,None,Y,#2 after 1s
TAKEPIC,C2,-,00:18.8,D850,1/1000,4.0,6400,0.0,RAW,None,Y,#3 after 1s
TAKEPIC,C2,-,00:18.2,D850,1/2000,5.6,6400,0.0,RAW,None,Y, #4 after 1s
LVPSTOP,C2,-,00:16.0,D850, , , , , , , ,stop live view, mirror down

 

My confusion comes from simply emulating Xavier's 'deluxe.txt' script that he reports to work with a card older than ours.  What's he doing that I am not?  I would like to cram bracketed exposures into the action packed times of Baily's beads and chromosphere.  But if compromise is needed, I will consider your burst approach with spacing between exposure changes.  

 

Please let me know what you observe when you try this script.  Meanwhile, I'll try yours.  I feel much better with two heads on this problem and will share all I find.

 

Best wishes