Messages

Faster Renderings: As well as optimising for multi-processors like above, when rendering make your render composition 'view' 1.5% in size and it will make more use of your RAM.

Its really tough stuff. Here's a 'good understanding' abstract for budding coders to look at where adaptive and non-adaptive techniques are currently employed.

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.122.2248&rep=rep1&type=pdf

bright3 better edge. check the curvature of the lamp. bright1 inconsistent. bright2 good balance.

Yep for me bright3 detail wins (see blind edges too). bright1 slightly better in highlight detail - see blind chord. (right click images into new window to analyse). bright2 somewhere in the middle.

yes.

a1ex, have you tried sampling more than 1 frame (i.e. two) during the interpolation and if so were there no averaging improvements? I know it would be more taxing but Id be interested to know what your findings were.

How about frame x {1 second in} (settled) - do the math - return to frame 0 / begin?

a1ex: In the dual ISO raw2dng interpolation, what ways are you looking at improving the accuracy of returned values and reducing artefacts? Ah I see you're already answering this.

A look at the artefacts of each of the test interp modes. Same dual ISO DNG opened 'As Shot - 5500 / Tint +17' (RawMagic and v0.13 OSX raw2dng no interp modes and opened with a tint of +66 then changed to +17).

https://www.dropbox.com/s/a3wd7s9sh5ud2sv/interp_artifact_test.png

median6 v mean23 v shadow25 v shadow 50 v shadow 100

https://www.dropbox.com/s/noa57vx3h6g7r0y/shadow25_50_100test.png

I really like shadow 50 together with the less purple artifacting in mean23

https://www.dropbox.com/s/2h7x59girnbxdpu/mean23_median6_shadow100.png

a1ex: Not at home right now so can't send you original raws. Here's my latest screenshots shoot produced today showing the processing of the same RAW file fed thru each of the methods: mean23 v median6 v shadow100.

https://www.dropbox.com/sh/4rpnyayri6n1jrk/2iMOz1n5Nl/mean2rb3g%20450%20avg%20hot%20pixels%20per%20frame.jpg

https://www.dropbox.com/sh/4rpnyayri6n1jrk/osCxpEq1xz/median6-1500%20avg%20frame%20hot%20pixels.jpg

mean23 averaging 450 hot pixels per frame.

median6 1500 per frame.

Shadow100 averages 1000 hot pixels per frame on the same tested RAW file.

EDIT: UPDATED (now with shadow100) picture showing the major differences of each against each other:
https://www.dropbox.com/s/2h7x59girnbxdpu/mean23_median6_shadow100.png

As to your A,B,C, question the grain in A to C goes from courser to finer chroma noise. Most people are going to choose the aesthetically more pleasing C, but I'd probably hunch for B. There's a tad more shadow detail in A which could be processed in post.

Dual ISO artefacts with mean23 but much less than with standard current raw2dng.

Excellent work a1ex.

A nice input method at last.

In a word. Yes. The SD controller is not that great in most DSLRs of pre-last year. BMCC may well be using better memory controllers. A simple explanation is found here;-

http://www.tested.com/tech/photography/455053-how-sd-card-speeds-are-often-limited-slow-memory-controllers/

Can you confirm this workflow improves on this?

Fine with 5DmkIII - tested an ext monitor and atamos without problems during record and no effect on the raw file.

@a1ex when we use 25p rec modes could you lockdown 50 shutter as base target on set ETTR instead of 30? - Its a tad annoying for manual lenses in PAL land.

Reidy's alright - he just winds people up now and again.

But you gotta hand it to him - he works hard at helping people on his site. There's nothing stopping others doing what he does - the reality is few can be bothered/are able to do it. Where's my free copy Reidy?! 

Code Select Expand

Buffer size experiment
ML v2.3.NEXT.2013Jun20.5D3113, 1b42291d5085+ (unified) tip
Mode: movie 1920x1080 25p, Global Draw: ON
CF LEXAR ATA FLASH CARD LEXAR ATA FLASH CARD                   
10312704 852
15885312 812
80896 238
8949760 900
28661760 821
16264192 872
24521728 858
16424960 819
13148160 831
15619072 799
1658880 694
21001216 924
8548352 749
19760128 881
5965824 815
24906752 893
13968384 757
16770048 818
4542464 801
13063168 860
33263616 829
30312448 866
16713728 808
3887104 793
21841920 876
12846080 785
16192512 870
3333120 726
6719488 876
17313792 871
23374848 865
2865152 744
32488448 896
18588672 847
4186112 736
821248 672
3661824 845
18359296 814
19875840 922
22754304 814
13665280 775
24957952 863
28555264 826
33442816 824
10708992 833
20242432 794
8774656 853
11216896 809
23156736 873
12750848 840
12369920 824
2157568 805
15279104 879
29943808 872
30321664 857
19004416 803
268288 408
23778304 926
467968 496
1905664 740
32368640 843
19829760 761


Definite improvement a1ex, getting continuous at 24p/25p 1920x1152 now on the lexar 128Gb 1000x. Global Draw On, ETTR On

Quick tests:

24p - 1080 - 92.8MB/s - 10% idle continuous
24p - 1152 - 93MB/s - 5% idle continuous
24p - 1280 - 95.7MB/s - 1-7ms idle > 1000 frames stopped
30p - 1080  - 91.8MB/s - 1ms idle stopped at 502 frames
30p - 1.85:1 - 93.8MB/s - 1ms idle continuous
25p - 1080 - 91.9MB/s - 5% idle continuous
25p - 1152 - 92.2MB/s - 49ms climbing to 80ms idle continuous

Incred1ble work a1ex. :-)

Also getting around 25% improvement on slower 800x cards.

DNG files look good.

Thanks. Testing now...

@a1ex I want to test this on the 5DMKIII immediately - have you got a build link?

http://www.glyn.com/data/glyn/media/doc/ATP_IG_CFast_card_spec_v1.0.pdf

Is CFast technology worth the look?

Many of the latest builds I have here for those who need them. If ML want me to remove this link let me know.

https://www.dropbox.com/sh/357k2kgbf4dx2zy/Upam6H6urE

I'm feeling better all the time for buying a used Lexar 128GB 1000x from an ebay guy for £300. :-)