Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Markus

Pages: [1] 2 3 ... 14
1
Can you try this on 5D3?

2
@D_Odell
I always convert mounted dngs with slimraw and put them on disk. Working directly with mounted dngs isn't stable with MLVFS on Windows either. It was pretty stable and faster with pismo but no support except standard 14bit raw there and mlvfs has all metadata handling working correctly.

3
Wow, that's a lot of dng files :).
It sounds about right that indexing would take this long. I'm even surprised you got all working in the end.
I would also suggest exporting to dng files directly through mlv_dump using the -c option which also compresses the dng lossless. Nothing against slimraw(really cool and fast) but you wouldn't need to go through mlvfs.
I wonder how batch_mlv will work on a humongous mlv file.

After checking all the files from the live event where we filmed ~20h 12bit lossless raw there is one file that's only outputting 50K frames when It should have 80K+ in it. I know this because the filesize and number of frames that I get out of It does not add up. I can also see the missing gap in the timeline I made In premiere with all cameras synced up from the event.
I guess It's a metadata corruption where the file thinks it only have 50K frames in it even though there is much more data there. Does anyone have any experience with this kind of problem? Is it possible to get the missing files out of the mlv or is this a lost cause?

Got the same amount of frames both from MLVFS and mlv dump/mlv batch, both missing 30K frames...

4
Just wanted to share my latest experience dealing with large amounts of data on Windows 10 using bouncyballs x86 compile of MLVFS with support for lossless.

We filmed a live show with a runtime of around 2h with 5 cameras. We filmed the show twice the same day and collected over 3tb of 12bit raw lossless data.
You can film around 93min with sound on one 256GB CF card and it turns out that one Original LP-E6 has just enough power to power the camera for that amount of time.

Had no big problems filming but the postprocessing was extremely slow. The idx-file generation on on these big and long file must have taken 24h or more. I just left my office computer on and checked in on it remotely from time to time.
After all idx files had been generated I started converting all mounted dngs through slimraw to disk. Strange thing here was that mlvfs stopped serving files after converting around 170K frames. So remotely I had to remount and restart Slimraw at 170K intervals to get through the 1.5 million frames (thank god for remote desktop :-)). After that everything went smooth exporting DNXHD files from resolve.

Is the slow experience with MLVFS  consistent with using it on Mac or Linux?

/end rant

5
General Development / Re: Silent (touch) controls for 5D Mark III
« on: April 09, 2018, 02:19:31 PM »
Metering timer, great tip!
If the faked MPU message works direct silent control sounds very useful aswell!

6
General Development / Re: Silent (touch) controls for 5D Mark III
« on: April 09, 2018, 12:20:29 PM »
Oki, maybe I just missed that you can set ISO control to the silent touch buttons. I only film raw though.

The problem I have with iso control is that the thumbwheel isn't active all the time. Canon locks it if it's inactive for a few seconds. So you always have to press halfshutter to make it active before turning the wheel to make your iso adjustment. I'm searching for a way to have direct input of ISO like you have of the of the aperture with your long finger. That total control is really useful for documentary and live event work.

Sorry if this a little bit off topic.

7
General Development / Re: Silent (touch) controls for 5D Mark III
« on: April 09, 2018, 10:58:14 AM »
Could it change ISO?  I have ISO set to rotating thumbweel on my 5D3. Maybe just tapping silent would be nicer. How Canon locks the thumbwheel so you have to half press the shutter to unlock it again is coubersome and prone to bugs like the "busy bug" that locks/blocks all exposure change. Or maybe there is a way to stop canon from locking thumbwheel in the first place, so that it's active all the time :-).

8
I said too much to fast, takes some time to load them when the files are large. When I tried to mount 2TB+ files at the same time Mlvfs got stuck on loading forever. I'm going to try to add one big file at a time to the mount directory to see if It has a easier time indexing them one by one.

I'm using a x86 version of mlvfs I got from bouncyball that has support for compressed mlvs. Latest Dokan version.

9
Thanks clanlee for detailed instructions and links for MLVFS win 10 x64 Install!
Got it running now. I had to go in and manually delete all old dokan entrys in regedit to be able to install new dokan. Even though I had uninstalled previous versions there were so much crap left there so the new dokan thought old versions were still installed and refused to install at first.

Mounting some really large files now, several around 200GB. It seems to take quite awile for dokan/mlvfs to load them. Is this normal? Previous versions i think it was pretty much intant even if the files were very large?

Also, does this version support mlvs with lossless compression? All the files I'm testing to mount now are 12bit lossless.

Switched back to old version with support för lossless. Files mounted instantaneously and it still worked with new version of Dokan.

10
Reverse Engineering / Re: UHS-I / SD cards investigation
« on: April 06, 2018, 01:05:11 PM »
Trying to activate brute force mode for testning on the Sandisk Extreme pro 512GB bit it does not seem to react to me pushing halfshutter when the test starts?

I think i accedently got it runnig the first time i tried this module but now it just does the normal tests.

11
Reverse Engineering / Re: UHS-I / SD cards investigation
« on: April 05, 2018, 08:54:03 PM »
I can try that tomorrow. Would be awsome to get it to work on this card.  8) :D

12
Reverse Engineering / Re: UHS-I / SD cards investigation
« on: April 05, 2018, 07:16:04 PM »
Quote
@Markus
Are you running the test in photo mode as when I tried it in video mode I got loads of crazy stuff coming up.

Just a thought..

I got a little bit faster results on the modes that worked but error on the same ones as when I tested in video-mode.

13
Reverse Engineering / Re: UHS-I / SD cards investigation
« on: April 05, 2018, 06:12:24 PM »
No more luck with the 512Gb Sandisk Extreme Pro with fw 113.
Same errors.  :'(

===================
2018/04/05 17:02:51
===================
Before the hack: r:20MB/s w:20MB/s  W:20MB/s R:21MB/s  8)  [best 20MB/s]
SDR50 @ 96MHz  : r:36MB/s w:35MB/s  W:34MB/s R:39MB/s  8)  [best 35MB/s]
SDR50 @ 96MHz  : r:35MB/s w:35MB/s  W:35MB/s R:39MB/s  :)  [best 35MB/s]
SDR50 @ 80MHz  : r:31MB/s w:31MB/s  W:30MB/s R:34MB/s  meh [best 35MB/s]
SDR50 @ 80MHz  : r:31MB/s w:31MB/s  W:30MB/s R:33MB/s  meh [best 35MB/s]
SDR50 @ 120MHz : r:err [SAFE]  [BACK]
SDR50 @ 120MHz : r:err [SAFE]  [BACK]
SDR104 @ 96MHz : D0 D0 r:36MB/s w:35MB/s  W:35MB/s R:39MB/s  ::) [best 35MB/s]
SDR104 @ 96MHz : D1 D1 r:36MB/s w:35MB/s  W:35MB/s R:39MB/s  ::) [best 35MB/s]
SDR104 @ 80MHz : D0 D0 r:31MB/s w:31MB/s  W:30MB/s R:33MB/s  meh [best 35MB/s]
SDR104 @ 80MHz : D1 D1 r:31MB/s w:30MB/s  W:29MB/s R:34MB/s  meh [best 35MB/s]
SDR104 @ 120MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 120MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 132MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 132MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3
Best: D0 D0 r:36MB/s w:36MB/s  W:35MB/s R:39MB/s  :)  [best 36MB/s]
Best: D0 D0 r:36MB/s w:33MB/s  W:35MB/s R:38MB/s  ::) [best 36MB/s]

14
Reverse Engineering / Re: UHS-I / SD cards investigation
« on: April 05, 2018, 04:44:06 PM »
I'll Downgrade to 113 test it there.

15
Reverse Engineering / Re: UHS-I / SD cards investigation
« on: April 05, 2018, 01:25:00 PM »
Ok, that fixed it.
write 35mb / read 39 Best
Not much luck on higher speeds. Any idea why?

5D3 fw123 "4K raw video recording; lossless compression" version of ML

CrystalDiskMark bench for reference:
Seq Read: 81.47MB/s Write 77.89MB/s

===================
2018/04/05 12:15:25
===================
Before the hack: r:20MB/s w:20MB/s  W:20MB/s R:21MB/s  8)  [best 20MB/s]
SDR50 @ 96MHz  : r:35MB/s w:34MB/s  W:35MB/s R:38MB/s  8)  [best 34MB/s]
SDR50 @ 96MHz  : r:35MB/s w:35MB/s  W:35MB/s R:38MB/s  :)  [best 35MB/s]
SDR50 @ 80MHz  : r:31MB/s w:30MB/s  W:30MB/s R:33MB/s  meh [best 35MB/s]
SDR50 @ 80MHz  : r:30MB/s w:30MB/s  W:30MB/s R:33MB/s  meh [best 35MB/s]
SDR50 @ 120MHz : r:err [SAFE]  [BACK]
SDR50 @ 120MHz : r:err [SAFE]  [BACK]
SDR104 @ 96MHz : D0 D0 r:35MB/s w:34MB/s  W:35MB/s R:38MB/s  ::) [best 35MB/s]
SDR104 @ 96MHz : D1 D1 r:34MB/s w:34MB/s  W:35MB/s R:39MB/s  ::) [best 35MB/s]
SDR104 @ 80MHz : D0 D0 r:30MB/s w:30MB/s  W:30MB/s R:33MB/s  meh [best 35MB/s]
SDR104 @ 80MHz : D1 D1 r:30MB/s w:30MB/s  W:30MB/s R:33MB/s  meh [best 35MB/s]
SDR104 @ 120MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 120MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 132MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 132MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3
SDR104 @ 160MHz: D0 D0 r:err [SAFE] D0  [BACK] D0
SDR104 @ 160MHz: D1 D1 r:err [SAFE] D0  [BACK] D1
SDR104 @ 160MHz: D2 D2 r:err [SAFE] D0  [BACK] D2
SDR104 @ 160MHz: D3 D3 r:err [SAFE] D0  [BACK] D3

Best: D0 D0 r:35MB/s w:34MB/s  W:34MB/s R:38MB/s  ::) [best 35MB/s]
Best: D0 D0 r:35MB/s w:34MB/s  W:35MB/s R:39MB/s  ::) [best 35MB/s]

-----------------------------------------------------------------------
CrystalDiskMark 5.1.0 x64
-----------------------------------------------------------------------
* MB/s = 1,000,000 bytes/s [SATA/600 = 600,000,000 bytes/s]
* KB = 1000 bytes, KiB = 1024 bytes

Sequential Read (Q= 32,T= 1) :    81.474 MB/s
Sequential Write (Q= 32,T= 1) :    77.886 MB/s



16
Reverse Engineering / Re: UHS-I / SD cards investigation
« on: April 05, 2018, 12:27:49 PM »
Just tried this with Sandisk extreme Pro SD 512GB rated at 95/mb sec but I'm just getting "malloc error". The camera can see the 512Gb card fine but seems like the module cant handle the big card? Any ideas? Also noticed that the camera could not format the card even though it can see it.

5D3 FW 123

Would be really cool to get this working with this card =).

17
Ok, thanks!

18
Trying to get mlvfz working for compressed mlvs. I have normal mlvfs working on this computer but when issuing mount command with mlvfz instead of mlvfs.exe nothing happens. Is their some step of the install process I'm missing/ have forgotten?

Mlvfz works fine on my other computer...

Win10

Any help is appreciated.

19
@Danne: that sounds sweet, gonna try it out!
:-)

20
Thanks for reply Danne & masc!

@Danne: Switch seems awsome but still on Windows here though :-).

21
So you can use this app now for DF averaging and put out new cdng files on Win (or maybe soon)?
If that's so I have to try this out!

@Danne, You have a program for osx that does this aswell?

22
I'll try, thanks for your help :) If I'll find what's causing the problems I will report it here in case someone else has it too. And the x86 version works fine. I deleted the html files while testing out different versions of mlvfs and forgot to add them again... So I can at least test it now.
I might have had the same ptoblem. In one of my office computers my antivirus program somehow made mlvfs crash the fist time I installed it, after that no amount of reinstallation could resolve this problem. Installing mlvfs on my gaming PC avoiding the antivirus program to interfere I got it working there. It's like something permanently broke in the system I first tried to install it on. Haven't tried reinstalling win due to the pain of it.

Skickat från min SM-G900F via Tapatalk


23
Using 256gb CF cards and lossless compression gives you 80min+ rec times on one card. Really good if you wanna film documentary or interviews in raw! A couple of 256gb cards and you're set for a whole day of shooting :-)
Mk3 ML raw + da vinci resolve is awesome!

24
Raw Video / Re: Uncompressed 14-bit RAW video testing - 5D Mark III
« on: January 16, 2018, 05:36:54 PM »
Raw is Intensive on your recording media not on the cpu. Recording compressed video data is more process intensive = hot cpu (though the camera deals with this with dedicated chips that are tasked to handle the compression processing)

25
4K is overrated, higher bitdepth raw, dynamic range and lowlight performance is worth so much more. 2k Raw = sweetspot res vs filesize compromise.

Pages: [1] 2 3 ... 14