A3000IS poring thread - page 13 - DryOS Development - CHDK Forum  
« previous next »
Pages: 1 ... 8 9 10 11 12 [13] 14 15   Go Down

A3000IS poring thread

  • 142 Replies
  • 56405 Views
*

Offline cagiva

  • *
  • 33
Re: A3000IS poring thread
« Reply #120 on: 11 / January / 2019, 17:47:07 »
Advertisements
@reyalp
Yes, I used two different SD cards in the 100d camera to run the tests.  One 32GB and also an 8GB card.
« Last Edit: 11 / January / 2019, 17:51:19 by cagiva »
Logged
*

Online reyalp

  • ******
  • 14080
Re: A3000IS poring thread
« Reply #121 on: 12 / January / 2019, 21:29:28 »
Quote from: srsa_4c on 11 / January / 2019, 13:13:53
This is getting really odd. The failures remain, only 3 shots are made instead of 5.
...
Anyone else have an idea what could cause this?
This is pretty weird.
The 100c and 100d capt_seq and boot look essentially identical, so it seems unlikely to be a porting difference.

The error is timeout, which would usually mean the hook isn't reached, but maybe it really is just taking too long?

Looking at hooktest, the timeouts are 5 sec, which seems a bit short. The first one works, but the second fails. Given long exposure and raw, 5s between the raw hook and the next shot could be marginal. So maybe the hooktest errors are spurious and unrelated to the kap_uav error.

@cagiva:
Is the "CHDK raw" option set in the script menu? It should have defaulted to off.
Was it dark when you tested?

Attached is a version of hooktest with the hook wait timeout increased to 10s. Please run it on the 100d camera and post the log. A log from the 100c camera wouldn't hurt either.
Logged
Don't forget what the H stands for.
*

Offline cagiva

  • *
  • 33
Re: A3000IS poring thread
« Reply #122 on: 12 / January / 2019, 22:05:40 »
Quote from: reyalp on 12 / January / 2019, 21:29:28
Is the "CHDK raw" option set in the script menu? It should have defaulted to off.
Was it dark when you tested?

Hi @reyalp,

Yes, the "CHDK raw" option is turned Off.

When I ran the tests yesterday, it was during the day. However, I was indoors with natural sunlight.

It's now night time here is California, but I ran your script twice indoors in a room with fluorescent lights and the test Passed successfully (log attached).  I can run it again outside tomorrow morning and post that log.
Logged
*

Online reyalp

  • ******
  • 14080
Re: A3000IS poring thread
« Reply #123 on: 12 / January / 2019, 22:42:02 »
Quote from: cagiva on 12 / January / 2019, 22:05:40
When I ran the tests yesterday, it was during the day. However, I was indoors with natural sunlight.

It's now night time here is California, but I ran your script twice indoors in a room with fluorescent lights and the test Passed successfully (log attached).  I can run it again outside tomorrow morning and post that log.
No need. In the last log, I can see it waited ~5.5s where it would have timed out at 5 before:
Code: [Select]
0021730: exp=1967 dir=A/DCIM/128___01 hook_shoot wait
0027220: exp=1967 dir=A/DCIM/128___01 hook_shoot ready
So I'd say this mystery is solved, but it's not clear what's happening with kap_uav. I haven't looked at the details of that issue, but if this camera/card is exceptionally slow, it's possible that script could be affected too.

I'm surprised it takes this long with raw off, that seems pretty slow. For the most part, these cameras don't take more than a couple seconds to shoot a jpeg.

If you have shot review enabled in the Canon firmware, that might explain it.

Just for the record, was this srsa_4c's build, or the autobuild?

You could also run the CHDK benchmark and post the log here. Go to CHDK misc menu, tools, make sure Log and SD Test are checked. Don't bother with this if shot review was on.
Logged
Don't forget what the H stands for.
*

Offline cagiva

  • *
  • 33
Re: A3000IS poring thread
« Reply #124 on: 12 / January / 2019, 23:27:56 »
Quote from: reyalp on 12 / January / 2019, 22:42:02
If you have shot review enabled in the Canon firmware, that might explain it.

Just for the record, was this srsa_4c's build, or the autobuild?

You could also run the CHDK benchmark and post the log here. Go to CHDK misc menu, tools, make sure Log and SD Test are checked. Don't bother with this if shot review was on.

@reyalp,
Shot Review is disabled.

I'm using the autobuild 5150

As per your request, I ran the CHDK benchmark with Log and SD Test checked.  Here is the log file.  Thanks!
Logged
*

Online reyalp

  • ******
  • 14080
Re: A3000IS poring thread
« Reply #125 on: 12 / January / 2019, 23:36:32 »
Quote from: cagiva on 12 / January / 2019, 23:27:56
Shot Review is disabled.
Darn, that seemed like a good theory  :(

Quote
As per your request, I ran the CHDK benchmark with Log and SD Test checked.  Here is the log file. 
SD speed is fairly slow (5-7 MB/s), but that doesn't seem slow enough to explain the problem.
Logged
Don't forget what the H stands for.
*

Offline cagiva

  • *
  • 33
Re: A3000IS poring thread
« Reply #126 on: 13 / January / 2019, 00:02:33 »
I also ran the CHDK benchmark and the hooktest script you sent early, on the other camera with the 100c firmware; which has a faster SD card.  Here are the log files, just in case this might help.
Logged
*

Online reyalp

  • ******
  • 14080
Re: A3000IS poring thread
« Reply #127 on: 13 / January / 2019, 17:16:34 »
Quote from: cagiva on 13 / January / 2019, 00:02:33
I also ran the CHDK benchmark and the hooktest script you sent early, on the other camera with the 100c firmware; which has a faster SD card.  Here are the log files, just in case this might help.
Thanks. Benchmark shows the faster write speed 6-9 MB/s, and the hook wait times are a bit under 5 sec, so that is consistent with the other results.

I don't really understand though why it would be ~5s between the end of one shot (start of raw hook, nothing saved to card yet) and start of the next (remote hook) with half press held. A3000 is a low end cam, but I would expect it to be capable of at least something like 0.5 FPS using this method. The Canon continuous spec is 0.8 FPS.

D10, elph130 and elph180 all take 0.3 - 1s at this point, so 5+ seems like a real outlier. Not sure what this means...

edit:
@waterwingz original diagnosis (https://chdk.setepontos.com/index.php?topic=10822.msg139057#msg139057) seems consistent with this: The camera seems to be actually taking ~5s to shoot when it should take ~1.
Since the 100c camera exhibits similar behavior (just slightly under 5 rather than slightly over), it's probably not a hardware failure.

@cagiva
Can you try shooting in continuous mode manually, and check whether you get roughly 1 shot per second, or one every 5 seconds? Make sure raw is disabled in CHDK.

If it's 5-ish seconds, try the same without CHDK running.

Running hooktest with continuous mode enabled in the canon firmware might also be a useful datapoint.
« Last Edit: 13 / January / 2019, 18:36:52 by reyalp »
Logged
Don't forget what the H stands for.
*

Offline cagiva

  • *
  • 33
Re: A3000IS poring thread
« Reply #128 on: 14 / January / 2019, 04:06:51 »
Quote from: reyalp on 13 / January / 2019, 17:16:34
Can you try shooting in continuous mode manually, and check whether you get roughly 1 shot per second, or one every 5 seconds? Make sure raw is disabled in CHDK.

If it's 5-ish seconds, try the same without CHDK running.

Running hooktest with continuous mode enabled in the canon firmware might also be a useful datapoint.

 "Save RAW" is disabled under the RAW menu.

Shooting in continuous mode manually also takes ~5s per shot.

Without CHDK running (SD card unlocked), takes ~5s in continuous mode as well.

I ran hooktest with continuous mode enabled.  The first time, it reported some errors.  However, I ran it again and it didn't report any errors.  Just in case, I attached both log files.  Thanks.
Logged
*

Offline cagiva

  • *
  • 33
Re: A3000IS poring thread
« Reply #129 on: 14 / January / 2019, 13:12:51 »
I tried again this morning with sunlight; which seems to be a factor.  With CHDK running (SD card locked) and no script activated, the shooting in continuous mode manually takes ~1s.  Then I ran your latest hooktest script and the the shooting seems to be now faster in continuous and single shot mode as well (logs attached).

The forum only allows me to attach two files so let me run @waterwingz script and load that log in another post.

Bottom line is that the lighting seems to be a factor and this camera is apparently faster for continuous shots in outdoors.

Logged
Pages: 1 ... 8 9 10 11 12 [13] 14 15   Go Up
« previous next »
 

Related Topics