M9 Banding and noise issues are gone!

[k-hawinkler]

Advertisement (gone after registration)

Many thanks. Very interesting.

Somehow I can't find which firmware version you are using?

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[brianv]

Faster memory cards gain speed by doing more operations in parallel. Sandisk used the term "ESPP stands for Enhanced Super-Parallel Processing". This means more operations going on in the card, at the same time. To me it means: more power draw, more electronic emissions. The fastest card that I've used in the M9 is a 10x PNY 8GByte. Most of the cards that I use are 4x. I tend to be conservative.

 

The UHS-1 card caused problems on several of my cameras and computers, I never used it in the M9.

[tobey bilek]

The error can be in a card reader or readers cable to computer.

 

Lexar and Sandisk test the card for way more read/write cycles than you are likely to use in a lifetime. I buy the quality cards. The few dollars you are likely to save are not worth potential problems.

 

I also try to avoid downloading with a reader, use the supplied cable. Ripping the cards in/out of the camera connections can not be good for the camera or card. One can always get a new battery which is user replaceable.

 

Also do not leave the card plugged into the computer any longer than necessary.

 

I also buy new cards with a new camera, format them for that camera, and they stay that way until the camera dies. Never had any issues with cards.

[dasper]

Paul and Ron,

 

Thanks for doing these tests with the cards. I wanted to pick up some Transcend cards and a couple other various cards that caught my eye to test along with my various current cards. I have noticed banding under certain circumstances all throughout my M8 and M9 usage and would love to lessen it's effect.

 

Are you guys using the model TS32GSDHC10 cards or the model TS32GSDHC10E cards? The E is supposed to have additional error checking:

 

"Transcend's TS32GSDHC10E is a mid-range flash memory card with 32GB capacity, and almost identical to the TS32GSDHC10. The only difference is that this model, with the "E" on the end of the model name, has a built-in error reader that can detect if a corruption occurred during transfer, and then attempt to fix it."

 

Strangely the Amazon description for the TS32GSDHC10E mentions no error checking but the description for the regular TS32GSDHC10 does mention error checking:

 

"Built-in error correction(ECC) automatically detects and fixes transfer errors"

 

There are also 2 versions of the Transcend 16GB Class 10 card as well as the class Class 6 cards, one with an E on the end of the name and one without...

 

Would be cool to find out who is using what and be able to compare and see if there is any difference and also to try to figure out the flash and controller manufactures from different series and different brands.

 

Cheers

[rscheffler]

Many thanks. Very interesting.

Somehow I can't find which firmware version you are using?

 

I knew I forgot to mention something... I vaguely made mention to 'not having updated' on the first page. It's 1.176. I've held off on the update because I kind of don't want the camera suddenly shutting down, in case I really need to squeeze more use out of it, but I think there was something else that 1.196 mucked up, wasn't there?

 

Which reminds me, I forgot to mention the tests I did were with a freshly charged battery.

 

Paul and Ron,

 

Are you guys using the model TS32GSDHC10 cards or the model TS32GSDHC10E cards? The E is supposed to have additional error checking:

 

Dasper: it's the version without E. I didn't realize there are two versions - thanks for that info, though I cannot find reference to it in the Transcend website

 

As for which brand to pick... this is so subjective and anecdotal. The reason I picked Transcend was because I've used their CF cards in DSLRs since 2002 and never had a problem, so decided to stick with them for SD as well. There have been so many apparent compatibility issues with Sandisk cards in the M9 that I'm simply avoiding them.

[pico]

I think you have a ropey sensor and that changing the card has just lessened the problem..... again probably all down to voltage issues or some sort of interference in the sensor cells/channels.

 

The host, which is the Leica camera, is supposed to check what voltage the card requires and if it cannot provide it, then no processing occurs. Initial voltage requirement is 2.7-3.6V. If the Leica does not check this, then the Leica firmware is faulty. If the card lies, then all bets are off. I find it more likely that the Leica does its check properly and the card lies, even if inadvertently, because cards are mass produced, have less quality control per unit. Additional evidence that the Leica is correct and the card is not is that the Leica works with other cards!

 

You impressionists should look to the SD card standards.

.

[tele_player]

Advertisement (gone after registration)

You looked at the actual standard from sdcard.org?

 

-Robert

 

You impressionists should look to the SD card standards.

.

[brianv]

Many of the vendors are not reading the specs. The Sandisk UHS-1 card, fresh from Sandisk as a replacement of a defective card, could not be used after formatting with the utility from SDCARD.ORG. The one and only operation done with the card. Sandisk sent another.

 

Some cards emit more RF than others. This can cause problems with systems converting analog signals to digital. Use cards that emit less RF, less to creep into the analog side of the electronics.

[pico]

You looked at the actual standard from sdcard.org?

 

-Robert

 

Of course I have. It was part of my job.

 

And you?

[brianv]

What would be required is to run custom firmware on the camera to exercise the various portions of the electronics while acquiring images. Using a spectrum analyzer while running the diagnostics would be of value. Doing a Fourier analysis of the banding on the CCD compared with the Spectrum Analyzer should indicate if RF emissions are the culprit. Monitoring voltage levels on the supply side of the electronics should indicate if power draw is a problem. Anyway, that worked 30 years ago to find a similar problem with a Digital Infrared Sensor.

 

Or the OP could just stick with the cards that work well with his camera.

 

Put a short wave radio next to some kids toys running microcontrollers sometime. I had a TI SR-56 programmable calculator that made it's own music on any radio in the apartment when in college. Took it a few days to solve some problems.

[k-hawinkler]

Hi Paul,

 

Please, let me come back to the banding issue.

Your examples were given here http://www.l-camera-forum.com/leica-forum/2193044-post64.html.

 

I have been lately shooting my M9 via an adapter on the Leica APO-TELYT-R 1:4/280.

Somebody suggested light leaks might cause reflections in my system.

http://www.l-camera-forum.com/leica-forum/leica-collectors-historica/252227-leica-serie-5-5-ndx1-13026-a.html

I didn't think so, but checked it out anyway.

 

BTW, with top quality adapters from Novoflex, focusing is no guess work or trial and error at all.

http://www.l-camera-forum.com/leica-forum/digital-forum/256040-focus-apo-telyt-r-280-4-a.html

I simply compose and focus the lens with my NEX-5N or NEX-7, then swap the camera and adapter with the M9 and another appropriate adapter.

The lens stays in focus. No problem at all, just inconvenient. Well, I am anticipating the new Leica M and getting ready for it.

 

So, here goes.

I shot the image with M9 + Novoflex M-R Adapter + APO 280f4 + Lens Cap at 32 s, ISO=160, f=4, focused beyond infinity.

The raw image was then processed in three different ways in PS CS6, complete image scaled down to 960x639 pixel^2:

 

 

Contrast=-100 Highlights=+100 Shadows=+100 Blacks=+100 Clarity=+100 Exposure=0 Whites=0

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Contrast=-100 Highlights=+100 Shadows=+100 Blacks=+100 Clarity=+100 Exposure=+5 Whites=0

 

Contrast=-100 Highlights=+100 Shadows=+100 Blacks=+100 Clarity=+100 Exposure=+5 Whites=+100

 

The memory card used was Panasonic 16GB SDHC Memory Card Class 10 UHS-I

 

Panasonic 16GB SDHC Memory Card Class 10 UHS-I RPSDB16GB1K B&H

 

Thank you for coming up with investigating a memory card in the M9 in this fashion.

[brianv]

Paul's banding issue was external to the CCD, and ran across both halves of the segmented detector. The artifacts shown here do not all (some seem to carry at different intensity) go across the two halves of the segmented detector and seem unique to each half. Detectors have non-uniformity in response, and it is native to each detector. I think you are seeing mostly the non-uniformity, blown up to proportions that are visible. Try the same test with an older, slower card. See if anything changes.

[Paul J]

Are you guys using the model TS32GSDHC10 cards or the model TS32GSDHC10E cards? The E is supposed to have additional error checking

 

Hi Dasper, Unfortunately I'm not able to tell. I no longer have the packaging and mine doesn't say otherwise on the card itself.

[Paul J]

I have been lately shooting my M9 via an adapter on the Leica APO-TELYT-R 1:4/280.

Somebody suggested light leaks might cause reflections in my system.

http://www.l-camera-forum.com/leica-forum/leica-collectors-historica/252227-leica-serie-5-5-ndx1-13026-a.html

I didn't think so, but checked it out anyway.

Thank you for coming up with investigating a memory card in the M9 in this fashion.

 

Hi KH, thanks for sharing the findings. Really interesting test with the moon as well.

 

The test is also a good way to find out exactly what's happening with dust on your sensor. Instead of taking a shot with the lens cap on take a well exposed one of a white piece of paper with the lens stoped right down. Open in photoshop and in Levels press Auto levels.