Is one shorter camera-card-slot pin an issue for RAW recording?

Started by adrjork, October 01, 2019, 04:59:49 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

adrjork

Hi everyone,
inserting a brand new CF card into my 5D3 I noticed it was somehow hard/tight to put it into the camera-card-slot.
After removing the card, I noticed that one pin (in the camera-card-slot) was a bit shorter than the others. Probably it was caused by inserting the card in.
Now, my question: is that shortened pin an issue for RAW recording? And most important: is there a method/software to verify if pins working properly? (since by eyes I don't notice anything wrong in the footage...)
Thanks really a lot.

ilia3101

some pins are of different lengths, and some cards are harder to insert than others

my 5D2 has a short pin, I think you don't ghave any issues

Luther

If you have access to other 5D3 with ML, there's the "card benchmarking". It will measure R/W speed. Put the same card on both cameras and then compare the values...

Walter Schulz

Pins should not be of different length and operation may suffer (or not) depending on pin's designation in short/medium/long time (after additonal wear and tear) or never. There are pins transfering picture data and there are other pins controlling card's status/data transfer and there are power supply pins.
Maybe some info about which pin is affected would be helpful...

Worst case: Card slot replacement is not unheard of. Any soldering wizard able to fix a modern mobile phone will do it with reasonable costs.

adrjork

Quote from: Walter Schulz on October 01, 2019, 07:35:17 PMMaybe some info about which pin is affected would be helpful...
Thanks for your reply, Walter. I have two 5D3 and I find out surprisingly that both have the same issue at exactly the same pins:
on both, if you think the slot vertically, in the first pins-couple on top I have right-pin higher and left-pin lower, while in the last pins-couple on bottom I have right-pin lower and left-pin higher.
In your opinion, in which case should I understand that my card slot has to be replaced?
Thanks a lot.

ilia3101

I think it's supposed to be that way, both my 5D2 and card reader have a short pin at one end. It might be the power pin, so that it only connects once all the others are connected.

Walter Schulz

Sorry, I gave wrong information! Pins of a CF connector should indeed vary in length (according to specs, for example 2.0):
1, 13, 38, 50 (power pins): 5 mm
25, 26 (detect): 3.5 mm
all other: 4.25 mm

adrjork

Quote from: Walter Schulz on October 01, 2019, 11:36:44 PM
Sorry, I gave wrong information! Pins of a CF connector should indeed vary in length...
That's a good new Walter!
Quote from: Luther on October 01, 2019, 05:54:34 PM
If you have access to other 5D3 with ML, there's the "card benchmarking"...
I tried benchmarking on both my two 5D3s and the results are very similar but strange: in video mode both give me write speed around 78 MB/s and read speed around 110 MB/s, while in photo mode both give me write speed around 118 MB/s and read speed around 153 MB/s.
I wrote it's strage because my card is a 160 MB/s Sandisk Extreme Pro, and in video mode I can easily record 1980*1152 RAW14bit (that should be around 95 MB/s) endlessly. So I can't understand the 78 MB/s in video mode from benchmark...
Another thing: I wrote that the results from the two cameras are similar, but NOT identical, and it's strange because I used the same card for the test...

a1ex

Quotein video mode I can easily record 1980*1152 RAW14bit (that should be around 95 MB/s) endlessly. So I can't understand the 78 MB/s in video mode from benchmark...

These speeds are actually expected. While recording raw video, we are slowing down or stopping some non-essential functions in Canon firmware (labeled "small hacks" in mlv_lite) in order to squeeze a little extra speed. Disable that option, and write speed should be a lot closer to benchmark figures.

Quoteresults from the two cameras are similar, but NOT identical

If you perform the test 100 times, swapping the card between the two cameras every now and then, and you plot the results, you may notice the difference is not statistically significant. Refer to this thread for formulas and such (in particular, T-stat and P-stat, and this explanation).

Caveat: if you actually run the benchmark 100 times, you'll wear out the flash cells from your card. I've run similar tests with mine, so you don't have to ;)

adrjork