Leica M240 Wakeup Time

[jaapv]

Several thousand? What kind of card do you use? And what do you do if the card crashes? Believe me, it happens...

 

[ferider]

I haven't seen anyone mention formatting their card as exFat.

The fastest startup time I was able to achieve, from turning the camera on to being able to make an image with the M262, was 1.6 seconds. That was with a Panasonic Gold 8GB card.

Time of other cards I tested:

16GB Sandisk Extreme Pro (95MB/s): 1.8 seconds
16GB Panasonic Gold: 1.9 seconds
16GB Samsung Pro (white card): 1.75 seconds

In order to measure the time I turned on the camera while holding down the shutter button. I recorded this on video. The audio was imported into Audacity, which allowed me to "see" the sound of the on switch and end with the sound of the shutter opening. The audio spikes in Audacity allowed for a precise measurement.

Formatting the SD Card as exFat is the key. I also tried SD Card Formatter alone and in conjunction with formatting in the camera. That didn’t seem to make any difference. Once I formatted the cards as exFat I recorded my fastest start-to-shoot times. The 16GB Panasonic’s start-to-shoot time when the card was formatted only in the camera at the card's default MSDOS FAT format was 5.2 seconds. Formatting the card first in my computer as exFAT, then formatting the card in the camera reduced the startup time on this card to 1.9 seconds.

Try it on your M240's and see if the startup to shoot times drop below 2 seconds.

Thank you for that. Very interesting that formatting in camera retains the exFAT file system (just tried).

Roland.

 

[krötenblender]

Several thousand?

When I measured the about 12s from off to first shutter release, I had only 1900 RAWs on a 64GB card. But I sometime happen to have many more.

What kind of card do you use?

64GB and 128GB (class 10 SanDisk Extreme, Transcend SDXC UHS-I U3 64GB, Lexar Professional 128GB 1000x Speed SDXC UHS-II)

SanDisks are the slowest of these types, but they all differ only very little in that regard. Only when transferring to computer, the faster cards really show their speed.

And what do you do if the card crashes?

I'm taking another one out of my pocket. I have plenty of them... Also, of course, at least once a day (often more than that) I transfer all files to a computer, which in turn is saved on three hierarchical backup systems from hourly (TimeMachine on Mac), daily and weekly. - You know besides taking pictures, I'm a totally paranoid computer scientist regarding data safety. I had my experiences...

Believe me, it happens...

I know... That is why I only delete data, when necessary, i.e. when the card is full. Also I rotate SD cards every day and from time to time buy new ones.

 

[robert blu]

Not a 240 owner but interested to learn, if awaking time depends on the number of files on the cards how many should be the maximum in order to have a reasonable short waking up time?
I usually never have more than a few hundreds (200 jpgs+200 dng on my Leica x1) .

robert
PS: of course keeping other variables unchanged (same sd card)

 

[krötenblender]

Not a 240 owner but interested to learn, if awaking time depends on the number of files on the cards how many should be the maximum in order to have a reasonable short waking up time?
I usually never have more than a few hundreds (200 jpgs+200 dng on my Leica x1) .

A few hundred should not be a problem, if you can live with 1,5~3 seconds startup time.

Since creating real world test cases for this phenomenon is obviously very time consuming (and back breaking for the hardware...) I can not really prove, but I suspect, that the algorithmic problem behind the startup time has O(n^2).

What puzzles me, is why it depends on the number of files at all... Why does the camera need to read from the SD card, when starting up? Other digital cameras I checked, are independent from the number of files on the card.

 

[jaapv]

After problems with card compatability on the M9 Leica does a startup check of the card.

 

[krötenblender]

After problems with card compatability on the M9 Leica does a startup check of the card.

That makes sense. However, it is obviously not implemented very smart... It would be enough to check the file system struct and read the directory contents of the top level directory (finding DCIM/ ) instead of the directory containing the files. If this works, then there is no problem to read the files, since both (reading the directory struct and reading files) need the same type of low level operations of reading block data.

Also, I wonder, if this is not possible to do as a background task or in interrupt routines, as even the most basic micro controllers have the ability to execute some time slice or interrupt based multitasking (I work with lowest power µC in automotive - the M240 controller is like a mainframe compared to that, so I'm sure, the computational capacity is there).

 

[jaapv]

Well, the problem of the M9 was corrupted files, as I understand it because of a different interpretation of the SD standard by some SD card makers. That is the reason that Panasonic Gold cards work so well on the M9, the card controller does an internal read-write check, it seems.

 

[bobby_novatron]

Thanks for passing on the info about the exFat formatting, I'll have to try that.