Messages

Latest build:
...and yes I searched before asking.

Any chance someone could explain me why does the raw rec window shows I record at 94mb/s at 1920x1080, 1920x1152, 1920x1280 even though theoretically the recording speed is different (for example when I choose resolution i want to shoot on, it still shows that 1920x1080 will be recorded at 83mb/s and the rest res in different speeds). Thank you!

For 5DIII the maximum continous recording on my cards is 1920x1152 and I know a lot more users who can easily record in that resolutions.

UPDATE: I can confirm that with latest build I can easily record 1920x1280 continuous recording on 5DIII with the Hoodman Steel 64gb cards.

Hey Scrax! How are you buddy? Just tested the latest version with 32gb file and this is what I got:

raw2dng converter GUI for OsX
Beta ver.0.11

M05-1921 File Supported
Generating jpg preview
dyld: Library not loaded: /usr/local/lib/libexiv2.10.dylib
  Referenced from: /Users/KMA_WWC/Desktop/RAWToDNG/raw2dng.app/Contents/Resources/ufraw-batch
  Reason: image not found
/Users/KMA_WWC/Desktop/RAWToDNG/raw2dng.app/Contents/Resources/script: line 31:  1674 Trace/BPT trap: 5       "$UFRAW" --out-type=jpg --out-path=../"$FILE_NAME"_jpg *.dng >> "$LOG"
Generating avi preview
dyld: Library not loaded: /usr/local/Cellar/ffmpeg/1.1.3/lib/libavdevice.54.dylib
  Referenced from: /Users/KMA_WWC/Desktop/RAWToDNG/raw2dng.app/Contents/Resources/ffmpeg
  Reason: image not found
/Users/KMA_WWC/Desktop/RAWToDNG/raw2dng.app/Contents/Resources/script: line 31:  1676 Trace/BPT trap: 5       "$FFMPEG" -i ../"$FILE_NAME"_jpg/%6d.jpg -vcodec mjpeg -qscale 1 ../"$FILE_NAME".avi >> "$LOG"

Is this normal? It also created txt log file inside the dng folder, and also created separate empty jpeg folder. Not sure if its ok or not, but just wanted to report it.

Hi there! You are not the first person to experience that problem. Check here http://www.magiclantern.fm/forum/index.php?topic=6145.0

Please could you PM me and describe what exact steps you took? Thanks.

Hey KMA_WWC I apologize for sounding aggressive towards you earlier. Hope you find a solution and share it. Also, Canon 5D mark III formats exfat automatically if the camera recognize a card equal or larger than 128GB.

LB

No worries at all! I do understand the frustration of waiting for the new card, receiving it, trying to use it and at the end of the day it turns out to be just a piece of plastic. Then you have to go through the hassle of returning it, waiting for refund then ordering new one and so one and so one.

Yes you are right about on camera exfat formatting on MarkIII, but the thing is, I wouldn't recommend anyone to buy the 128gb Komputerbay card since its very rare to get a fast one. Most of the 128gb can't even pull off 75mb/s writing which means no 1080p continues recording.

It's great for other cards though such as Lexar.

After today's testing it looks like exfat for MarkIII for Mac users is the best way to go. Did multiple speed, recording, transfer, formatting tests today and didn't encounter any problems like I did with the other cards (which were fat32 formatted in camera).

Tomorrow I hope to do some tests with 5DIII and 5DII on PC and Mac with exfat . And also, I will probably do tests with fat32 format (even if the cards die, I will be ok with that. At least it will prove my and techs thoughts). There is a slight chance that with latest build things may work differently and Mac issues won't be an issue anymore.

Basically the issue with the cards right now isn't even on camera formatting, but the Mac itself. Their cards more likely will fail and get corrupted while being connected to the Mac computer than in camera on the shoot. That's all because of fat32 format which Mac (especially on with Komputerbay cards due to some way of internal processing) doesn't like at all.

I just did the 5 minute benchmark test on the 5D3, I thought I was using the may 28th build, -- but on the benchmark test it said
"ML v2.3.NEXT.2013.MAY18.5D3113"--Anyhoo :

My write times are:  ( I left out the read speeds) komputerbay 64 1000x card

What do you think?  return the card?

67.2  (buffer=2040k)
57.5   (buffer=1953k)
71.0   (buffer=3072k)
61.0  (buffer=2920k)
70.9   (buffer=4096k)
67.4  (buffer=3906k)
74.6  (buffer=16384k)
74.1 (buffer=15625k)
28.3 (buffer= 128k)

Try with the latest build (june 4) and do the same test. If nothing changes then I think you should return the card.

I've got a 5d2 and a 64GB 1000x Komputerbay, how do I make the card bootable after formatting on Windows?

Did you have any problems recording raw with the cards being formatted in fat32 on windows?

I'm not in any way trying to upset you or anyone else but a dead card is a useless card and should be sent back. A half dead card is also a useless card and should also be sent back. I got my card today and spent time trying to figure out how to work it.

Do you mean to say you got 4x komputerbay cards today and all 4 faild? Why in the world would you even recommend these cards?

Those 4 unusable cards failed 3 weeks ago, and I didn't say I recommended their cards, did I? I'm just trying to help new and current users of the Komputerbay cards. Today I received 32gb and 64gb cards from Komputerbay for testing. I will use their solution and see if its any good. If yes, I will post later my findings.

Check here my first findings about their cards and other http://www.magiclantern.fm/forum/index.php?topic=5391.0

Ok, problem with my card is that it works in my camera but not in my computer. I have formated the card in my camera. I can shoot on the card in the camera, the camera reads and writes just fine.

Your conclusion is to never format the card in camera?

If you are a Mac user then yes it's the best way to keep the life of the card. That's why I'm doing tests. There might be a better way. Just give me time. I just received the cards today.

Keep in mind that many people use exfat and they are not formatting their cards in the camera because they want to have the feature of recording single files bigger than 4gb. Those people have no complains since there are no other way. Maybe ML will create an exfat format feature in the camera in the future (most unlikely).

UPD: I think this thread would be better suited in the hardware section.

Komputerbay cards are as I understand really Lexar cards that did not pass Lexars quality control. That is basically from my understanding why there are so much variations with these cards regarding read and write speeds.

Also. My Lexar 1000x card formats and works perfectly in camera format and in computer format.

So from my point of view. If your komputerbay card works fine, then good on you, I'm really happy for you. But if it fails like mine did. I am f***ing lucky I didn't use that card on a job.

I'm just like you an unfortunate user (4x bad komputerbay cards). I don't use them on a pro work. I use my Hoodman Steel 1000x cards. But, at the same time I want to help people, and trying to find a way how they can prevent card failure. Komputerbay cards are getting more and more popular due to their cheap price and there needs to be a way to fix current issues. You can't just go and tell everyone don't buy their cards.

Am I right in that if I format the card in computer to exFat and then put the card in to the camera, and then format the card in camera, only the files will be deleted, no actual format is done in the camera?

I get failed format in computer when trying exFat and Fat32 (have tried on several computers and in several card readers) gonna send this card back to the store.

Have tried low-level format to. No success.

If you format it in camera, it will delete the files and format it back to fat32 even if it was exfat. I'm not 100% sure, but your card may not be workable anymore. So, I would say sending it back is the best thing to do.

What I'm trying to achieve with my tests is to find the most optimal way of using Komputerbay cards, so people have as less problems and corrupted/failed cards as possible.

[DISCLAIMER: Neither Komputerbay or I take any responsibilities for anything that goes wrong with you cards, camera, computer or any other device during and after the tests. Please do it on your own risk!!!]

Sorry forgot to mention that its better to format to exFAT for Mac users. The latest ML builds allow the use of exFAT formatted cards for recording single files for more than 4gb. You will have to format your cards from your computer. If you do in camera then it will be formatted back to the fat32.

I'm not 100% sure about MkII and how it behaves with the exFAT, komputerbay cards, and latest builds since I haven't done any tests with it yet (only on MKIII). I know that MKII is limited to 65mb/s writing speed, and the module may not support exfat. I will be doing tests with MkII though, I just don't have the camera right now.

If you would like to help! Follow this steps and do w/r 5min speed test, record something (record one file >4gb and one file <4gb), and transfer it your station. Repeat it a couple of times and please email me at khachikmkrtchyan(at)gmail(dot)com if you experience any speed drops, or issues with step by step explanation.

DISCLAIMER: Neither Komputerbay or I take any responsibilities for anything that goes wrong with you cards, camera, computer or any other device during and after the tests. Please do it on your own risk!!!

"Procedure for using the hack on 64GB card:

1. Format the card in Mac using exFAT (not FAT32 as the MAC default's to FAT32)
2. Make it bootable
3, Install the Hack
4. Insert the card in your Camera
5. Clear the buffer for your Camera

Start using the device.

If you use FAT32 - then the card most likely will slow down or corrupt the card (slow down on Mark III and corrupt on 50D/Mark II)."

If you are using MKIII then step 2 and 3 isn't necessary, since the hack is likely to be installed on SD card.

There is a solution to all your problems guys (it's a common problem with Komputerbay cards). I will post all my findings and results + with some insight info from Komputerbay tech/engineers, which should help people who bought their Komputerbay cards to get max performance and life time. It will be a big post explaining everything about KB cards from A to B.

I will post it as soon as I finish all my tests and find solution with KB. They might even come on board and help users with their cards, but that's not 100%!

One advice though for right now! Try not to format their cards to fat32 and keep it exfat as it comes originally! It should help (doing test, but it technically should be one of the solutions). You can help me too with that. Let me know if it works for you better than fat32.

If your card is dead or corrupted, please contact Komputerbay directly or PM me explaining step by step what you did from start (taking card from the package) until the end (when the card failed/not working).

Your card may have died when you connected to Mac (that's not 100% either, need more test to determine how they behave with Mac. Exfat should be the fix, hopefully).

Thanks.

Best regards,
Khachik.

What did you expect? The Toshiba is about on par with the Lexar, there are no faster cards. 

Well...I'd like to argue about "there are no faster cards." Hoodman Steel 1000X 64gb pulls of solidly up to 100mb/s and most of the time bounces between 108-116mb/s. Just saying

Don't forget Hoodman Steel is the top performer too!

system disk is full  time to get that raid array. I recorded a 60Gb file on my 64Gb card and was going to stop recording @ 60Gb but it stopped automatically @ 60 and is 62Gb in length. Not sure if it's gonna be corrupted, just moving some files around first.

Haha, so basically you decided to take the test to another extreme level with recording 60gb file

Perfect! Mad props scrax and rodobot for getting this working. I did a batch with one file over 2Gb, it's good for any size now yeah?

We just need someone with the latest build (exfat record ready build) to shoot/test out files bigger than 10gb and that would be perfect.

Let me know if you want me to give it a try before making it public. I have lots 4gb+ files.

Scrax please check http://www.magiclantern.fm/forum/index.php?topic=5247.1375 post #1376.

The guy might have found a way to fix 2gb limit.

Here is my take on ML RAW firmware. PLease let me know what you think. #NonCatDogFlowersWindowVideo

http://vimeo.com/66555717

More info on workflow etc you can in the description.

Thanks.

Here is my take on ML 14bit RAW hack. Used 5dIII with Hoodman Steel 1000x 64gb card. Please let me know what you think. #NonCatDogFlowerWindowVideo

http://vimeo.com/66555717

Is the casing made of metal? If so, have you noticed any difference in terms of heat? (I imagine that a CF card in steel casing would dissipate heat better than one made from plastic)

I did btw, sorry completely forgot to mention that. It is much much cooler compare to all the cards, I would say not just simply cooler, but zero heat on it. At least I didn't feel it.

Wow, it overlaps perfectly with the other Lexar from Jason Montalvo.

Glad to hear that I will be getting 4 more 64gb Hoodman cards pretty soon. I can test those to see if there is any significant difference in writing speed between same cards. I know for sure that hoodman quality control/check is top notch, so I expect them to have relatively same speed with not much difference.

The only issue I see is that I still can't record continues 1920x1280 even though card tops out up to 116mb/s writing speed, and 1920x1280 requires only 98mb/s. I think it's comes more to firmware tweaking and getting max out of it or as much as possible. Basically, optimization.

Just finished the 5h+ test (until full charged battery died). I used Hoodman Steel 1000x 64gb card. Writing speed 95% of the time was over 100mb/s. It dropped under 100mb/s only when the screen went to sleep, but when I got it back, speed returned back to over 100mb/s. Not sure why it happened.

Here are the file/results:
https://www.dropbox.com/s/qr3kpxh3xtleukl/BENCH.LOG

Please do let me know if that helps or if you need any other tests. I would love to help as much as I can.

Ok, I started the test of the Hoodman Steel 1000x 64gb card (from 5min benchmark tests I get writing speed up to 116mb/s and reading up to 143mb/s). I will run the test until my battery dies and post the results here. Hope it helps