Messages

Simply put, it is Toshiba 1.8" hard drive 50 pin CF adapter -> CE / 40 pin ZIF IDE ribbon cable -> JMicron 20330 ATA 150MB/s to SATA solution -> mSATA SSD (pictured here a Samsung PM830 256GB version).

The assembly was not successfully powered up by a USB3 CF card reader in computer. Plugged into the 5D2, the "format CF card" options posed up initially, then turn into a "CF ERR" on the top plate LCD. The cable stays flush through the card slot.

The SSD had been partitioned as Master Boot Record and formatted as FAT32 by restoring from another SanDisk 1067x CF card. Works fully through a USB3 to SATA bridge, has sequential read / write speed unto 400MB/s.

Two possible reasons for it to not work fully:
1. The camera's card slot only supplying 3.3V instead of 5V (yet the bridge should operate on 3.3V as well, so it might just be the SSD)
2. One side of the CF connector is not perfectly aligned with the slot, so it might have a electrical connection issue.

All components should be searchable on Amazon. The price of the 256GB SSD plus all the adapters is close to 1/2 of SanDisk Extreme Pro 1067x 64GB, so probably worth the risk.

I did some research on yours. You are actually very close to the solution, the external power supply is what you are missing.

Also what you need to find is the power voltage select pins on the JM20330, the default is running under 3.3v which is obvious low for a SATA device. There is another version of this broad have a jumper or switch to select the power voltage. From your description, "cannot format" is probably because of under voltage since camera is already "found" there is something in its slot.

The CF slot should not be able to provide more than 500mA of current (According to Sandisk) and most SSD nowadays are using far more than 500mA, external power supply is highly recommended.



Later I found some cheaper and easier way to do a conversion. The parts are on the way, lets wait and see.

AFAIK UDMA-7 has never been ported to IDE/PATA hard drives ...
Really don't know if there are any PATA-SATA converters able "to speak" UDMA-7.

Good point, I went back to check the datasheet of the converter (JM20330)

http://www.users.on.net/~fzabkar/temp/JM20330_Spec_Rev.-2.3.pdf

There are descriptions about the Ultra DMA operation mode on page 19.
000: Device Mode 100MB/s.
001: Device Mode 133MB/s.
010: Device Mode 150MB/s. (default)
011: Reserved.
100: Host Mode 100MB/s.
101: Host Mode 133MB/s.
110: Host Mode 150MB/s.
111: Reserved.

By putting the jumper between these three pins, users are able to select the UDMA operation.
Good Point again, but no need to worry about.

I have a new Idea.

Since the CF slot is in the IDE standard, I decided to first convert the CF slot into a standard IDE 44 pin slot so I can attach IDE hard drives


After I have the IDE 44 Slot, I should be able to convert it into SATA (there are boards with good price on this).

I might have to make my own cable, since I am in China right now it should be easy to reach.
Let me try if this works.

Reply if you see any problems 

Quick notice that ML is developed and released by a group of people who are doing this voluntary. Don't urge them or keep asking the same questions like "when is xxx model going to have ML firmware" if you are not a part of the developers, they are nice and friendly already by giving out the source code and the usable files out.

I think it's about sata, not ata ... The last standard of sata is sata3 (750 MB/s)

Oh yeah lol, I miss understood. However there are many examples doing CFexpress to SATA on YouTube. Shouldn't be a big deal.

I just got reminded about CF-to-SATA thread. And I was going to post here that finally we will be able to attach HDDSSD to our cameras, as there are CFast to SATA adapters (CFast is in mostly a fancy connector for SATA).

And then I checked again and it's CFexpress  Didn't even notice that there's a new standard already. PCIe v3 x4 / NVME - I wonder if someone will make  m.2 NVME -> CFexpress adapter 
They are a couple of CFexpress -> (m.2 | pcie) adapters, and while I can't find any photo of "the other side" of pcbs, all PCIe traces seems to run directly between connectors.

Hold on, ATA133 which is the last version of IDE, only runs at maximum of 133mb/s. Attach to a SSD or HDD seems just save money for the captaincy?
The exposed socket the external power supply seems more dangerous and unstable xD. But it is a cool idea anyway and did you already tried that out? If not, I am glad to work on it.

Yeah fair enough! Makes sense. Keep the M2 for monitors, and the EOS M for the fun stuff 😉👍

You know how it's like these days with film-makers, we all have to have monitors on our rig irrespective if we use them or not. 🤣
Cheers for the reply and keep up the good work. I recently got a 5D2 but I'm still hyped with using the M & M2.

Na....this is not working either, I am buying a HDMI hacker to force EOSM2 output a 480p signal in order to decrease the usage of the CPU. Lets see what will it do.

Using a merged M2 build from Danne and DForts M2Builds:
https://drive.google.com/file/d/1o6e0LMn7r3bbNj84O06ThT3qezL14iEf/view

This is cool!
Let me try out for this
Thank you!!!

Never experienced bad frames before on the M2 with three different branded monitors I've used. Also never had issues with the current build I'm using for the M2. Works flawlessly. Great camera! 

How...? I have no trouble using the built-in monitor, but with a 1080p monitor attracted, problems are just all over the place. 

Thanks. Let's skip raw_slurp for now.
The other version should be working normally not getting into 1800x when selecting mcm rewired. If you can, please test that as well only for mcm rewired 1736x976
This version:
https://bitbucket.org/Dannephoto/magic-lantern/downloads/crop_rec_4k_mlv_snd_isogain_1x3_presets_EOSM2_dfort_2019Jul27.EOSM2103.zip

Another News is this is the stablest version I ever used.
With 1080p monitor attached and 1600x863@14-bit(not losses), it can recording normally without bad frames.
With inside monitor, basically everything are so far so good.

This is a tip for future developing (if it keeps), this version is on the right track.

The color science and the quality of the image is awesome !
Glad to see these wonders

with "crop_rec_4k_mlv_snd_isogain_1x3_presets_EOSM2_dfort_2019Jul27.EOSM2103.zip" I found that when an outer monitor attached, EOSM2 have the trouble on the bad(pink) frames. But this doesn't appears on the EOSM. The problem of the image buffer? PS: the monitor is running on 1080p

I had raw working exactly like i wanted and loved it...then i recorded stuff today and the files completely disappeared.  The space is 'used' on the card but i cannot seem to locate them.  Any recommendations on how to scan for the MLV files?  I was able to see them in the camera viewer before i pulled the card out but now i can't find a trace.

Look at the name of the file in the mlv_play, and search it in your computer

https://youtu.be/LgIO7bQvFo4

Just for fun lol.
Anamorphic feeling
ISO 6400 and the noise level is acceptable

If you are after reliable continuous AF you should not consider buying a EOS M or M2.

To my testing... EOS M2's continuous AF is usable on the object that does not move too much, a interview for example.
However the EOS M indeed, do not have a reliable continuous AF even if the object is not moving.

EOSM2's AF system has been improved to the level which is usable.

Green line? Take off bit depth in the crop sub menu  and keep aspect ratio 16:9. X3 crop works best for dual iso. No aliasing or artefacts. The only issue I have are the focus pixels. 3x3 won't cure them with both x3 & dual iso enabled...

Yes, the focus pixels is what I mean. Focus pixels fix seems can not work with dual iso?

https://youtu.be/ezeRCz9z7a4

I took another shoot to see if dual-iso works,
for this video I used 100/800 on the dual iso and get 14ev stops of DR
The result is amazing and beautiful, however the green line still exist.
I tried use my way which I mentioned before, it doesnt works. lol
Still, it is good to see this noise level and DR on EOSM2. Good job

Hm... but MLVApp won't import dng files.

Emmmmm...I can change setting to DNG-lossless in order to do that

MLVApp removes focus pixels before dual iso is processed.How you want to do that? Exported MLVs still have focus pixels.

Remove the F-D first and output the dng
Than use dual iso dump

Hi, I tried using Dual iso on the M2. What I found is the D-R increased from 10ev to 12ev with 100/400 option ON (which is exciting)
However, when I am using MLV App to process the video, I found these interesting lines:
https://youtu.be/B_HAzRzC_RE

I turned both fixing focus pixels and Dual iso options ON and these line appears. But if I only turn on the fixing focus pixels ON, there is not wired lines jumping anymore. I guess this because the MLV App first calculating the Dual ISO images than remove the focus pixels. I will trying remove the focus pixels first and output, than let MLV App calculating the Dual iso value, see what will happen.

Hi Danne,

I send you a personal email
please check

don't care about this floor, I wasted it

[shaky, sorry for that]

New test video on the dynamic range of the EOS M2 with 14 bit-losses.

Because I put it on my bike...so it is shaky, sorry for that

However it shows a excellent DR. BTW this video is shooting under AF.

Also thanks "masc" for links to video and "d_fort"'s fpm files

Shutter speed == 1/54
FPS == 24fps
Aperture == f8
Lens == 17-40 f4

https://youtu.be/Z3iMnC3orPY

And finally I have a question : Will it be possible once porting of the core (By core I mean the feature set that is available now) is stable and "over" to have the same recording modes as the original M (Like those shown bellow) ?

I tried to move the crop.mo from eosm's firmware to eosm2's. However it doesn't work. Since these two camera are very similar, I believe it will come very soon.

->

GREAT THANKS !!!!