Personal View site logo
'Apocalypse Now' Experimental Series 1 Thread - BOOM, Intravenus - cbrandin/driftwood AN Soft/Cinema
  • 1089 Replies sorted by
  • @jakepowell I am sorry to make it get confused, because I was mistaken in the matrix. :-(

    Since "TYPE-ZERO" released previously was not an original matrix of Chris strictly, I remade another setting as "TYPE-ZERO1" with the pure original matrix of Chris anew.

    http://personal-view.com/talks/discussion/comment/82763#Comment_82763

    "what can we expect to see ?"

    The result which Chris expected from the first in 24p should be obtained. :-)

  • Sir Driftwood, congrats on the very first and unique "flat-4" Intra settings of Apocalypse Now ! Will test it as soon as possible !

    +Enjoy your vacation, you deserved it like no one other :)

  • I just did a test with the DREWNET version and it didn't span on a Transcend 32gb card class 10. It didn't span in either 1080p 24p, 24H or 24L mode. In H mode it recorded around 8.5 min, in L mode it recorded just over 12 min.

    The 32gb Transcend does speeds around 13-14mbs write, while the 16gb version did around 14-15mbs. I thought I had tried L mode in the Cluster 6 and it spanned but now I'll need to test that again to be sure. If it won't work, then AN won't be for me, I need a patch that at least spans in 24L mode.

  • @CRFilms Ill have to test that as well since i too use some cheap Transcend C10 / 32Gig cards as backup. However, i have noticed not all transcends are equal. I have 2 of the 32G cards and one preforms slightly better then the other, both where purchased at the same time and from Amazon. I now use 2 - 32G SanDisk Extreme 45/mb and it smashes through all these patches with ease, spanning and everything. I'm currently getting great results with DREWnet on the SanDisks. The faster cards make your camera recording more responsive as well, the transcends would "lag" a bit when starting/stopping at higher bitrates.

  • Sir Driftwood, first thankyou for your constant endeavor's for the GH2 community. Second ive had a brief play with BOOM-Intra on a couple of FD lens stopped down for optimal sharpness. The bad: I saw some very minor aliasing for the 1st time on GH2 (the aliasing is not noticeable when 2 feet away for my pin-sharp imac screen). The good: These are simply the best images I have seen from my GH2, they have by far the smoothest gradiation of any setting I have used yet were not at all 'soft'. All my clips averaged 148Mbit aswell if thats any use.

  • I am having big problems with all the latest patches. In camera playback is not working half of the time and for me that is a big dealbreaker. I think the Cluster v6 DREWNET is the best stable patch of the bunch. Great quality and stellar stability in almost all modes.

  • i too have had a brief play with boom . funnily enough with canon FD's they seem to create a rather nice image :) not soft - but from what i gather boom isn't? and as @Jizzyjones stated lovely smooth gradation and colours too :) guna put it more to the test this evening as I'm putting an installation up atm, but i may do a few film clips of that which will be challenging- lots of textures and low light / silhouettes etc :) cheers though hope france treats u well

  • @Jizzyjones Don't forget to read my warning: Boom! is best only for mid to tight shots. As a quick checkoff, if the scene youre filming's datarate is under say 140mb you should be fully making use of the setting. Anything above and you are in the danger zone of possible artefacts. Boom! Is a pure over the top experimental setting which is brilliant for closeup skintones.

    @jakepowell thnks. Bordeaux countryside is lovely.

  • I read it, but im not gonna lie I have no clue what the hell this 444 business is nor will I try too its beyond my intelligence! I trust my eyes, 'Canis skintone soft' has been my go-to setting for EVERYTHING that includes wides aswell as close ups, as wrong as that may seem. The scenes I shot were super contrasty and detailed, to me it looks excellent especially wides as there you can appreciate the wonderful new gradiation. I look forward to a full release an hope you keep this matrix it is excellent!

  • @driftwood interestingly all my stuff on boom so far plays straight up from camera - no need to switch off to get it to play the latest clip filmed... they are short clips though and close ups of skins under indoor light (for which this works beautifully on btw ) its a small thing but its nice regardless, don't know if this is of any importance just thought i would mention...

  • Hey guys, I went camping over the weekend and made a little video using the Apocalypse Now hack. So far I really like it, let me know what you think.

  • Interesting in terms of spanning results on an Extreme 45 mB/s 32 gig card:

    AN GOP 6, 24L- could not span, maximum overall bit rate of fairly still footage of 50.4 mb/s

    AN 'IntraVenus' Settings - Intra GOP1, 24L - spanned fine on 2 different scenes, maximum overall bit rate of 75.4 mb/s

    Could be due to the higher GOP?

    Very new to all of this.

    Loved the quality of these patches- man, the noise is so much more controlled than other patches I've tried!

  • Valkyrie 444 TYPE-ZERO1 'Apocalypse Now' with cbrandin 'Pure Original 444 Matrix' by bkmcwd
    Film Mode:Nos -2,-2,0,0 (AWB:B+5)


    LUMIX G X VARIO 12-35mm F2.8
    LUMIX G 20mm F1.7
    LEICA DG MACRO ELMARIT 45mm F2.8
    Contax Plannar 50mm F1.4
    Contax Sonnar 85mm F2.8
    CANON New FD 50mm F1.4
    CANON New FD 35mm F2
    Asahi Pentax Super Takumar 55mm F1.8

  • @petersoncinema The video looks good. Any issue with noise at all?

    Is it me or do the images from these new patches make the GH2 look like Sony camera footage? Not saying it as a bad thing...just....something about it.

  • @petersoncinema Nice video! Which AN?

    @Driftwood Just a couple of observations. AN Drewnet soft does seem to handle problem colors better (red & green especially). It looks better to me in low light than CM Night. So far, it's handling 720p OK which is a must for me. It handles motion well.

    All these are more important to me than who's image of a 200% rivet shows the impact divot more clearly.

    One interesting thing which is neither good or bad, just different. The first thing I do when I import into AE is check the WB. Color Finesse recommends a cheat by just hitting auto color to WB then go in and adjust the image. This works 90% of the time but not in AN. The WB is off and the colors are deeply saturated.

    The only reason I mention this is that it works for all the settings I've tried except AN, so there is definitely something different with it.

    Overall...I love it!

  • @petersoncinema Cool vid, I might go do that some time!

  • @driftwood which setting is better for panny lenses 12-35 x and 14-140 g.

  • @Lambo nice boom test , can we see an ungraded version?

  • @rajamalik, test out the settings for yourself.

  • @jakepowell yes, but tomorrow now I'm tired!

  • Apocalypse Now "Nebula 444 Soft" from Abraham Aponte on Vimeo.

    Check out the link of sample footage of Nebula 444 Soft. this was done with no color grading at all, the settings i chose are described in the video enjoy this awesome patch. Idk how to embed the video to the forum correctly for you guys to see it from here any advice on how would be great!

  • Regarding spanning.

    As I have said many times before, the card that has the highest chance of spanning is the SanDisk 95MBs 64GB. Many testers use this to test new settings and it has demonstrated higher performance than any other card tested with the GH2 (as of a few months ago).

    If you are using that card, the overwhelming majority of settings will span in 24H. If a given setting does not, it is very notable and important to say so.

    There are other settings (both by @Driftwood and others like @LPowell and @Ralph_B) that place a higher priority on spanning and will span on a much wider variety of cards.

    The Transcend cards are among the least likely to span with high bitrate performance settings. The Delkin Elite 633x and the SanDisk Extreme cards rated at 30 or 45MBs will be able run a wider range. But the SanDisk 95MBs 64GB version will span with more settings than any other card tested to date. It is THE card to get for GH2 hacked settings at the moment.

    If you want to see one of the areas where these settings differ more, you can use controlled (and repeatable) lighting and position to create banding in your favorite old setting and then shoot it with one of the new matrices. The differences range from subtle to noticeable depending on the material and may be very important for some people.

    In regards to earlier comments about the diffrent settings and the difference in the way they looked vs the way that different cameras looked, I would offer my personal paradigm on the topic.

    Every scene has a certain potential range. No matter what camera or lens you bring in, all possible results will fall within that potential range. If you leave the lighting as is, and the scene is static, then the next variables are the lens and position from which the scene is shot.

    Once those parameters are locked off, "specific potential" for the shot is determined by the combination of lens and camera. No codec can exceed the specific potential of the shot, only try to live up to it. On a conceptual level, most people feel that a codec is doing a good job when it appears (to them) to get close to what they originally perceived or what they wanted to perceive. Since the original subject is infinite in detail and the the codec produces a finite approximation, something will always get lost.

    With higher bitrates, the potential for a codec to have enough data available to record a more accurate approximation of the subject increases. The more difficult the detail is to render in a small number of bytes, the more important the number of available byetes (or the compression approach) becomes.

    To perform a meaningful comparison of the limits of different codec settings, the shot must exhibit visual deficincies in at least one of the tested settings. If you are testing new settings, start with a scene - macroblocking, banding, jaggies or other undesirables - in your old settings. To get a better sense ofline, shoot as a JPEG at 1920x1080 (or whatever resolution is equivalent to your video mode) and RAW.

    That provides a baseline with which to compare the settings and a good indicator of whether the scene is sufficiently demanding to illustrate the differences with the naked eye.

    @cbrandin @lpowell I would skip using screenshots and just export specific frames as PNGs. This is easy to do with minimal processing in apps like VirtualDub for 8-bit files.

  • @Mrhechang

    Thanks for good looking sample footage, mate! Can the beauty of image quality be recognized also in YouTube? :-)

This topic is closed.
← All Discussions