I just finished "Last House on the Left" BD and thought the movie was outstanding. BUT - could someone tell me why there was so much grain in the darker scenes? It was so bad at times that the picture was completely obscured. FWIW, I've noticed this problem in the night shots in other BD movies and can't understand why the studios would release BD discs in which the night shots are so horrible. I actually own some of these offending BD titles in SD and have noticed that the SD night scenes are far superior to the BD photography, which seems to be the opposite of what one would expect!

Oh - I did search for an answer to this problem in other threads but did not find any information other than "possible artistic intent".

The Last House on the Left (2009) - Review of the BD.

The source is, as you would expect for a new release, pristine. Unfortunately, the post-processing on this transfer is way tweaked. I saw the film theatrically, and it had a much more natural, film-like look. Here, whites are way bumped to the point that contrast flattens. Blacks also could have been richer, which also dampens fine detail and depth. Sharpness is good, not great, while the color palette has an odd bluish edge to it that further undermines realism. Compression artifacts are not an issue, though there is a surprising amount of noise to the picture. If this is the filmmakers' vision than fair enough, but it feels noticeably different that the theatrical look, and given the intended gritty subject matter, the glossiness of this presentation does not do the film any favors.

http://bluray.highdefdigest.com/2365/lasthouseleft2009.html

I haven't seen this movie either in a theater or on Blu-ray, but it is not uncommon for scenes shot on photographic film in dim light, at night, or simulating night, to have more natural print grain than shots taken in bright daylight or under bright lights. It's part of the nature of the photographic film and the filming process. The extent of the grain also depends on the type of photographic film and camera used.

(If we're talking about digital, video shooting, all bets are off.)

Anyway, if the grain was not there in the theatrical release, it shouldn't be there in the disc copy, and it could actually be transfer noise. If the grain WAS there in the theatrical release, it should be there in the disc copy.

John

I haven't seen this movie either in a theater or on Blu-ray, but it is not uncommon for scenes shot on photographic film in dim light, at night, or simulating night, to have more natural print grain than shots taken in bright daylight or under bright lights. It's part of the nature of the photographic film and the filming process. The extent of the grain also depends on the type of photographic film and camera used.

(If we're talking about digital, video shooting, all bets are off.)

Anyway, if the grain was not there in the theatrical release, it shouldn't be there in the disc copy, and it could actually be transfer noise. If the grain WAS there in the theatrical release, it should be there in the disc copy.

John

AFAIK - when they shoot at night , they use a faster ASA film - like 400 instead of daytime shooting which may use 100 or less. What I do not know is if 400 is a granier stock than say 100.

AFAIK - when they shoot at night , they use a faster ASA film - like 400 instead of daytime shooting which may use 100 or less. What I do not know is if 400 is a granier stock than say 100.
That makes sense, but if this were true ALL movies would have grainy night shots, which is not true since we've all seen BD movies which beautiful night photography. Also, I've noticed that some of BD discs which have bad night scenes have good night scenes in their SD versions . . . which makes no sense to me.

That makes sense, but if this were true ALL movies would have grainy night shots, which is not true since we've all seen BD movies which beautiful night photography. Also, I've noticed that some of BD discs which have bad night scenes have good night scenes in their SD versions . . . which makes no sense to me.

This link says the higher the film speed - the more grain:

http://www.photonhead.com/beginners/filmspeed.php

There are so many variables when shooting film that we would need someone like Camman to join the discussion.

And we would need Mr. D to talk about post processing.

Maybe both will join the thread.

If you want good quality night scenes.

You lit the movie. If you have enough light, you can use a much less grainy filmstock.

I've noticed that a lot of my BluRay and HD DVD discs have what looks like compression artifacts in dark scenes.

I did a search and couldn't find anything about it so I figured I'd post a thread and see if anyone else has experienced this and if this is just a common thing or is it something that will hopefully be addressed with future Blu releases.

My equipment is as follows:

Sony VPL-VW60 projector
118" Carada screen
Pioneer VSX-92
PS3
Xbox HD-DVD add-on

That makes sense, but if this were true ALL movies would have grainy night shots, which is not true since we've all seen BD movies which beautiful night photography.
It depends on how much extra light they added. Thirty years ago there was no ISO 400 color film so they had to pump a lot of light into night shot to get decent exposure. These days film is more sensitive so they can use less and in some cases no extra lighting. Sometimes it works and there is little grain. Sometimes there's less light than they expected so they push process the film which generates more grain. In the end you have to either reshoot ($$$) or do with what footage you have.

There's a scene in My Year of Living Dangerously which shows how this works. They shot during a sunset. The shots where Mel Gibson is speaking look great. The shots of the guy he was talking to are extremely grainy. That's because they shot Mel Gibson's lines first, and shot the other guy's lines when the sun was much dimmer.

Unborn and Last House were both put out by Universal within a few months of each other. One looks like a top notch transfer (Unborn) where dark scenes exhibit grain, but it doesnt look bad by any means.... Then there's Last House which, within the first 5 minutes of the movie, you can clearly tell there's more than just grain in those scenes... Perhaps it's digial noise accompanied by macro-blocking, but it looks bad.

I'm sure there are factors involved, but bottom of the line is that Last House, as that review says, looks like it was digitally molested :)

BTW, another movie with natural grain, yet a decent amount of it -- 30 days of night. It's been a long time since i've seen it, but I remember it looking good, even though grainy.

Generally not a problem with the source, but an indication of an improperly calibrated display.

You need to have your display ISF calibrated, or if that is too rich for you, at least buy one of several available calibration discs, read and follow the instructions, and do it yourself.

The specific area of calibration that causes this effect is an incorrect (too high) black level.

Compression algorithms see no point in preserving data that is not going to be visible to the viewer on a properly calibrated display.

Vern

Unborn and Last House were both put out by Universal within a few months of each other. One looks like a top notch transfer (Unborn) where dark scenes exhibit grain, but it doesnt look bad by any means.... Then there's Last House which, within the first 5 minutes of the movie, you can clearly tell there's more than just grain in those scenes... Perhaps it's digial noise accompanied by macro-blocking, but it looks bad.

I'm sure there are factors involved, but bottom of the line is that Last House, as that review says, looks like it was digitally molested :)

BTW, another movie with natural grain, yet a decent amount of it -- 30 days of night. It's been a long time since i've seen it, but I remember it looking good, even though grainy.

Maybe the result you are seeing is due to the film process that was used to shoot the 2 films? I can't find anything on The Last House on the Left:

The Last House on the Left

http://www.imdb.com/title/tt0844708/technical

While the unborn was shot in 35mm Panavision

The Unborn

http://www.imdb.com/title/tt1139668/technical

Hope those discs aren't from Warner.:eek:

From Wikipedia, something further to think about:

"Day for night, also known as nuit américaine ('American night'), is the name of a cinematographic technique used to simulate a night scene. Mainly intended to avoid costly (and technically challenging) night filming, outside scenes can instead be shot during the day, using tungsten-balanced rather than daylight-balanced film stock or with special blue filters and also under-exposing the shot (usually in post-production) to create the illusion of darkness or moonlight. Lighting the characters two to three stops more than the background also helps the effect, but doing so requires powerful lights to compete with the sun. Many of the night scenes in the film 'Jaws,' for instance, were done this way, as was the rock-climbing sequence in 'Deliverance.' Historically, infrared movie film was used to achieve an equivalent look with black-and-white film.

While this technique has largely disappeared owing to advances in film technology, especially the use of digital intermediate post-processing, and increasing viewer expectations, it was recently used in the 2007 film '28 Weeks Later,' due to the impossibility of shooting in an entirely dark London, and in a dream sequence in 'Lord of War' for logistical reasons. It was also used in the 2008 film 'Mamma Mia! The Movie' for the opening scene.
Interior day-for-night shooting can be more time consuming and labor intensive. Grips need to cut all the daylight entering onto set. If the scene is 'blocked' or staged away from windows or other openings to the outside, the light may be simply blacked out with cloth or plastic sheeting. However when windows or doors are seen from camera, these openings must be 'tented' to allow some exterior dressing to be seen.

While never fully successful in creating 'realistic' night, the special visual style of day for night nowadays has many fans among historic-movie buffs due to its frequent use in early B-movies, Westerns, and film noir. Day-for-night shooting seems to have become more common in recent years, which goes against the trends of a decade ago."

John

Don't quote me, but I think it's a byproduct of MPEG-4. I've seen it a lot in large flat areas of color in animation and black areas in all mediums.

I guess the psychovisual algorithms say, hey, I can remove a lot of data from this area since it's all the same, and may go overboard.

Don't quote me.

Too late...

Blockiness in shadows is not video codec exclusive.

Blockiness in shadows is not video codec exclusive.

Correct
I you can have blockiness everywere else, there is no reason why you cant have it in the shadows.

It sound like a calibration problem (even if a block is an encoding artifact)


Also like to add that people focus to much about codecs they forget its just a tool for the encoder.

Its not the codec, its how you use it.

Don't quote me, but I think it's a byproduct of MPEG-4.
Oh no, another quote! If anything, I would say this is most prevalent in MPEG-2 titles but it seems to crop up in quite a few older and newer titles in a codec-independent manner.

In general, I agree with Vern Dias in that the OP's brightness and contrast settings may not be calibrated to crush black to the expected level. However, I disagree that this is not a problem with the source material since these artifacts do not need to be present regardless of the playback device's calibration settings which are user preferential and also influenced by the ambient viewing environment - strict ISF calibration standards are not the final word outside of the batcave.

Thanks for the replies everyone.

Actually my VW60 HAS been calibrated, and if the problem showed up on ALL of my Blu Ray or HD DVD discs I would tend to agree that perhaps there's a problem with the black levels... BUT since it doesn't show up on every Blu Ray or HD DVD disc I would have to suspect it's a difference with the encoding.

I was always under the impression that with the Blu Ray / HD DVD format they were allowed much more headroom with the bitrate --thus little to no compression, but I guess I was mistaken?

I'm just wondering if others are experiencing the same issues.

For instance several of the Harry Potter HD DVDs were pretty bad with the artifacting.

I was always under the impression that with the Blu Ray / HD DVD format they were allowed much more headroom with the bitrate --thus little to no compression, but I guess I was mistaken?

We still talking about 95% compression of the source.

Thanks for the replies everyone.

Actually my VW60 HAS been calibrated, and if the problem showed up on ALL of my Blu Ray or HD DVD discs I would tend to agree that perhaps there's a problem with the black levels... BUT since it doesn't show up on every Blu Ray or HD DVD disc I would have to suspect it's a difference with the encoding.

I was always under the impression that with the Blu Ray / HD DVD format they were allowed much more headroom with the bitrate --thus little to no compression, but I guess I was mistaken?

I'm just wondering if others are experiencing the same issues.

For instance several of the Harry Potter HD DVDs were pretty bad with the artifacting.

The video in these formats is lossy and compressed.

As others have said, your brightness setting in your display likely is set too high. Check a pluge pattern on a calibration disc to verify whether or not you are seeing blacker-than-black. Even pro calibrators will sometimes bump the brightness setting up a few notches after dialing in video black.

As others have said, your brightness setting in your display likely is set too high. Check a pluge pattern on a calibration disc to verify whether or not you are seeing blacker-than-black. Even pro calibrators will sometimes bump the brightness setting up a few notches after dialing in video black.

Improper gamma calibration will also make visible artifacts that should be hidden to the eye.

I remember that one of the times I changed projectors, I happened to watch the beginning of Charlie & the Chocolate Factory on HD DVD. During the opening credits, when the CG "camera" swoops down the smokestack, the image broke up into horrible blocky artifacts. I'd previously watched the disc on my earlier projector and not seen anything like it. Turned out that I had to go into the new projector's service menu to change the gamma setting from "video" (which was too boosted) to "film". The problem immediately went away.

We still talking about 95% compression of the source.
Depends on the studio. :D
Seriously, if you look at the real compression concerning loss of information and compare losslessy compressed to Blu Ray then 15 Mbit/s AVC is about a factor of 10 or less and 35 Mbit/s is about 4 or less. Visually quite close to indistinguishable in real time.

You can tweak codec settings to improve the quality in dark regions.

Two quotes; awesome.

One of the places this phenomenon stands out for me was in Sleeping Beauty, when the fairies are giving gifts to the princess and the scene dissolves into a spiraling galaxy. There's blocking everywhere in the space.

Sounds more like a temporal artifact rather than a spatial one like it sounds the OP is describing.

I was always under the impression that with the Blu Ray / HD DVD format they were allowed much more headroom with the bitrate --thus little to no compression, but I guess I was mistaken?

Well.. at 24fps and 8bit/colour the lossless video rate would be over 1GIGAbit per second.. so at 20-30MEGAbit per second theres a lot to throw out.

Might go check out the Harry Potter HD-DVDs and have a look-see..

Ah ok, I didn't realize it was that much.

I loaded up my Avia disc and the blacks were indeed a tad too high. So that has helped, but it's still apparent in many scenes.

Specifically in Harry Potter Goblet of Fire HDDVD.

Well.. at 24fps and 8bit/colour the lossless video rate would be over 1GIGAbit per second.. so at 20-30MEGAbit per second theres a lot to throw out.

The losslessly compressed rate is relevant for actual information loss. Not the uncompressed rate.

Depends on the studio. :D
Seriously, if you look at the real compression concerning loss of information and compare losslessy compressed to Blu Ray then 15 Mbit/s AVC is about a factor of 10 or less and 35 Mbit/s is about 4 or less. Visually quite close to indistinguishable in real time.

And were did you find this data?

And were did you find this data?
Estimates from experiments with x264 AVC encoder which has a lossless mode. You can roughly compress 10 times losslessly with += grainy material is what I found so far. Single frames get you about a factor of 3, adding temporal another factor of 3. I don't know what the extremes are with very noisy material or completely clean material. Maybe some people here have experience? The problem is you need uncompressed originals or you don't start with the full information content and compress earlier compression artifacts together with the already prefiltered real detail.
Red One Red Ray codec claims 10 Mbits/s 4K material looking about as good as Red Code at 28 Mbytes/s. Of course Red Code is prefiltered ~3.2K real luminance detail and they use wavelets so their numbers look far better than what one would expect at first sight coming from 1080p with AVC or VC-1.

The losslessly compressed rate is relevant for actual information loss. Not the uncompressed rate.

True..

Estimates from experiments with x264 AVC encoder which has a lossless mode. You can roughly compress 10 times losslessly with += grainy material is what I found so far. Single frames get you about a factor of 3, adding temporal another factor of 3.

That's pretty cool.. so are the bit rates you're seeing around the 100mbit/s mark? Have you tried any other lossless encoders?

True..
That's pretty cool.. so are the bit rates you're seeing around the 100mbit/s mark? Have you tried any other lossless encoders?
100-150 or so with a grainy source. No other encoder tested.

Also, I've noticed that some of BD discs which have bad night scenes have good night scenes in their SD versions . . . which makes no sense to me.

Makes no sense to me either, since I often see the opposite: if anything night scenes on Blu Ray tend to be smoother and more solid.

Details in dark areas of the screen seemed to be harder to render on the compression used for DVDs, and often I would see a lot more "noise" and macroblocking in dark areas. (Sometimes this could have due to the fact that darker scenes would be filmed with higher speed film, which is grainier, and grain is tougher to compress).

So on the whole I find the dark scenes in many hi-def transfers quite a bit better than in SD DVDs.

So on the whole I find the dark scenes in many hi-def transfers quite a bit better than in SD DVDs.
I agree, but there have been isolated instances where the night scenes in a BD movie have not been as good as the SD version of the same movie. One title that I recall had this problem was "Full Metal Jacket". The first high def release was so bad that they eventually came out with a "Deluxe" version which simply corrected the problems with the night shots. Another title in which the BD version night scenes weren't as good as the SD movie was the original "Stargate" movie.

Seeing compression artifacts in dark areas is not surprising. It is one of the most challenging situations for the encoder (not the codec) to deal with. The eye is surprisingly good at seeing artifacts there, and the encoder surprisingly blind to it!

In the glory days of VC-1 encoder development, we did a ton of work in detecting such problems and dealing with it. It was a major area of focus for us and as encodes went on, there were fewer and fewer areas like this.

And unless something has changed, encoding shops like Warner GDMX used to use displays with elevated levels as to see artifacts in dark areas. I don't think many others did this though.

Please, don't say that there are VC-1 titles from Warner with artifacts in dark areas :). I am sure there are some. I am just stating the goals and progress in that area, not an absolute solution.

As an aside, I am at CEDIA and was just walking by a BD movie being played at one of the booths on a flat panel with elevated levels (and turned up sharpness) and the blocking artifacts in the dark area were just incredible. No doubt calibrating the display would help but there were far more artifacts than I expected to see.

Amir, is Microsoft still giving out free copies of PEP?

I think it all depends on what you as an individual consider bad. I might look at the same movie and think wow that looks lovely and look at the detail. Others might look at it and say why is there grain in that scene. Night photography does tend to have some extra grain especially on older films but it's hard to tell if this is objectionable without having seen the films you think are bad.

I have heard some display technologies such as Plasma can make grain actually seem bad. Not sure if thats true but i have read it. Others can comment about that.

I guess the bottom line is that on SD DVD many movies were filtered and robbed of their grain structure due to the limitations of the technology at that time but with Blu Ray they can replicate the cinema experience and give us the film look which often means there is a grain structure which is sometimes very fine and sometimes thicker depending on the film stock, camera lenses, filters and time of day the scene was shot.

So you have an option to keep the film look intact or degrain the night time scenes. Doing the latter could result in a loss of detail if they use their tools aggressively and that has happened on some good releases.

I think we need to hear about more movies you have watched where you find the grain a distraction so we can understand if it's the film or whether it's a poor compression job and just noise and not actual film grain or whether it's your display which is enhancing the grain and making it stand out. Do you watch with your display calibrated and sharpness turned down or sharpness turned up and is the contrast set correctly ?

I've noticed that a lot of my BluRay and HD DVD discs have what looks like compression artifacts in dark scenes.

I did a search and couldn't find anything about it so I figured I'd post a thread and see if anyone else has experienced this and if this is just a common thing or is it something that will hopefully be addressed with future Blu releases.

My equipment is as follows:

Sony VPL-VW60 projectorI don't know about the VW60, but I had a VW80 and it had a problem displaying 2 steps from video 16 to video 18. I could not get video 16 and video 17 to ever look any different than each other. For instance, on something like the Spears and Munsil test disc the checkerboard with 16 and 17 on the first pattern would never appear. I believe this (along with video 18 maybe being too bright compared to video 16 anyway) caused problems in some dark images like in AVP:Requiem where I would see an ugly step down low or ugly transitions from black to a little above black. If the person doing the compression was seeing 2 steps from 16 to 18 and that is shown as one step instead then something that looked like a fairly smooth transition to them can become an ugly transition.

If you have the AVP:Requiem disc then I think chapter 7 or 8 has a good example. It is the one with the bums underground. When the woman comes down there is a shot where the Alien is just about to attack her where she is on the left and the Alien on the right. Part of the image on the right is video 16 I believe and then transitions to brighter as it gets closer to her. That transition looked worse on the VW80 than on some other projectors I have looked at it with. The steps from 16 to 18 can be tested with the AVS Blu-ray test pattern disc or the Spears and Munsil disc.

--Darin

This link says the higher the film speed - the more grain:

http://www.photonhead.com/beginners/filmspeed.php

There are so many variables when shooting film that we would need someone like Camman to join the discussion.

And we would need Mr. D to talk about post processing.

Maybe both will join the thread.

Today's 500 asa stocks have amazingly fine grain given the sensitivity. Its been a while sonce I worked on a film where anyone bothered to shoot slower than 400-500asa. When the newer Vision stocks came online from Kodak a few years back I had to recheck camera information as I didn't believe the grain structure I was looking at was 400asa: it was about the same as 100 from the mid 90s.

Its customary to over-expose film slightly ( up to a stop in some cases) to push the darker ranges up past the toe of the film's response curve and then print back down to restore contrast. Negative captures about 2.5 stops of headroom in the whites that gets softclipped off on print so you have a huge amount of latitude to print back down and still preserve white details ( one of the things that massively differs between film and video/digital).

The result of this is that you record more variation towards black and have less grain. 99% of the imagery I deal with straight off the neg is overexposed to a certain degree. Its very rare for me to be asked to grade a plate up the way ( when I am its usually to fix something that someone else has messed up and there is no alternative).

However film sets are not very controlled environments ( generally slightly organized chaos even on big budget films) so its not difficult to imagine a DOP making a decision that impacts the blacks at to preserve some upper ranges in the film.

DI grading can make a very big impact in how grain appears in the blacks ( previously grading was limited to additive offsets (printer lights) . Nowadays the grading tends to be applied as fairly arbitrary RGBHSI curves and the image can be further isolated with secondary mattes and keying. I've seen imagery where almost the entire latitude of the neg was curved around into the print range so almost everything visible on the neg is translated to the print.

Some DI graders are better than others. I've often had to take DI shots into VFX land to better adjust the levels with all sorts of strange bizarre keying and matting processes. Generally we get called in when a DI grader is having problems that would resolve as noise where he has crushed the image too far or built too steep or crunchy a curve : so its possible that some shots that may have benefited from a bit more refined treatment maybe didn't have the budget or time requirements to get passed into a VFX process.

I've even had to comp in reconstructed clean plates ( mini-mattepaintings) to give better range in certain areas.

Then of course the shot may be a day for night. I often find directors are willing to live with a bit of compromise in the noise levels if the overall grade is more convincing. (the opening of Mama Mia for example)

I have heard some display technologies such as Plasma can make grain actually seem bad. Not sure if thats true but i have read it. Others can comment about that.



LCD exagerates grain way more than plasma ( plasma generally doesn't have enough precision near black to make it any more useful than as a secondary client monitor though).

Managing grain when viewing an LCD takes a bit of getting used to if you come from a CRT. (Grain itself is a bit of a black art : all the preset film stock models in any package I've seen are very far from the mark as to what real grain looks like: the better tools let you crank it around by eye to get a good match : even then you normally have to render it out to confirm it as the software viewers in most apps usually aren't 100% accurate).

I can usually get a good match in less than 10 minutes but I've been doing it for 12 years. I often see people fiddling for hours just to match grain....very disheartening if I've trained them:(

LCD exagerates grain way more than plasma ( plasma generally doesn't have enough precision near black to make it any more useful than as a secondary client monitor though).



Thats interesting.

Lets say i'm watching on an LCD projector does that mean the grain levels are not right and if so in what way would i notice this onscreen or would i even notice it even if the projector is calibrated ?

I can usually get a good match in less than 10 minutes but I've been doing it for 12 years. I often see people fiddling for hours just to match grain....very disheartening if I've trained them:(
Why is it that you need to fiddle with parameters like that? Are the software guys just lazy or is there a deeper reason that the software analysing the reference grain in the images would have problems to find the correct parameters itself?

Negative captures about 2.5 stops of headroom in the whites that gets softclipped off on print so you have a huge amount of latitude to print back down and still preserve white details ( one of the things that massively differs between film and video/digital).

But the digital competition is getting closer. I have seen the latest ASC/BSC camera test and the highlight detail was quite close with the best digital cameras even in shots designed to show film's superiority there. Shadow detail was often better than on the film. Next year's improved sensors from ARRI and RED will close the gap further.

Why is it that you need to fiddle with parameters like that? Are the software guys just lazy or is there a deeper reason that the software analysing the reference grain in the images would have problems to find the correct parameters itself?

Getting access to the actual mechanics that play a role in the look of film grain is difficult for a start because the film manufacturer's generally don't let people snoop around their lovingly crafted clever bits of high technology (film).

Emulsion properties improve and change even with the same types of film stock over time. Throw into that exposure variances, development variances and it makes it a very hard thing to pin down to an easily modifiable set of criteria that are effective and up to date.

The nasty film grain tools just apply differing levels of noise to the various color channels, the better ones take into account luminance and saturation levels to better try and replicate how grain appears in bright vs darker areas.

Then you have the fact that something like film grain tends to render slightly differently depending on whether its a preview on the screen or an actual baked out render.

The best film grain tool is still the one from Cineon ( the preset stocks are innacurate but they always were) as it allows you to tweak grain meaningfully . Kodak built the seed pattern with an emulation of the t-grain technology prevalent in modern film stocks. The film grain tools in shake and nuke are direct lifts from this ( Kodak made the cineon source code available). These work fine but again are almost identical in terms of application to the cineon one.

Its outdated but gives you the right controls to eyeball match the grain ( assuming you know to factor in the differences between how it will look on a display and actually back on film).

I've yet to see an automated system that gets close enough to be honest: Shake has one that is better left unused IMO. However the tools built using the cineon tech are pretty much all you need to get a good grain match and as it only takes 10minutes for a decent comper to nail a match its not really worth forking out for some expensive state of the art system that doesn't work that well anyway.

But the digital competition is getting closer. I have seen the latest ASC/BSC camera test and the highlight detail was quite close with the best digital cameras even in shots designed to show film's superiority there. Shadow detail was often better than on the film. Next year's improved sensors from ARRI and RED will close the gap further.

I've yet to see them get good shadow detail and filmic headroom in the whites within the same shot.

I actually think the sony implimentations with a sensible custom curve on the camera get about the closest to film . RED is still pretty nasty in my book ( had a play with some footage the other day).

I've yet to see them get good shadow detail and filmic headroom in the whites within the same shot.

Yes, thats' where the difference still is visible. If you want good detail and roll off at both ends at the same time with high contrast shots. And no excessive noise plus accurate colors.

I actually think the sony implimentations with a sensible custom curve on the camera get about the closest to film . RED is still pretty nasty in my book ( had a play with some footage the other day).
The new ARRI sensor has far less noise than the best Sony when shooting the same dark scenery. The comparison was brutal. The current RED is a generation behind. And for best results you need to grade from the RAW data (not 10 Bit DPX) and use the latest firmware (20 or 21). The new RED sensors should be in the same ballpark as the new ARRI sensor. Interesting times ahead, especially for regular consumers who can afford a Scarlet (3000$).

I find the color on the RED to be one of its biggest weaknesses . It all seems to be reds and cyans and little else... even RAW from the camera looks severely lacking in gamut ( quite unpleasant the way it renders colors).

Its sharp enough to rival film in most cases but I'm always very concious of the latitude and color limitations.

I'm not saying the sony solutions are technically better but they seem to have come up with a better compromise in terms of the cameras abilities whilst maintaining good latitude and color.

This article is interesting for anyone who wants to see what areas digital have the hardest time competing with film. Very telling.

http://www.24p.com/ASC_Pictures.htm

This article is interesting for anyone who wants to see what areas digital have the hardest time competing with film. Very telling.

http://www.24p.com/ASC_Pictures.htm

And bare in mind the film had been converted to video so it will not fully represent the latitude actually available from the original negative.

The DX100 is not exactly a pro level camera either though.

And bare in mind the film had been converted to video so it will not fully represent the latitude actually available from the original negative.

The DX100 is not exactly a pro level camera either though.

Nope alot has happend since then. And it was just SD for all 3.

But i think its shows very clear that improvments in latitude is more important then just improvments in resolution.

I find the color on the RED to be one of its biggest weaknesses . It all seems to be reds and cyans and little else... even RAW from the camera looks severely lacking in gamut ( quite unpleasant the way it renders colors).

Did you use the new firmwares with the new color science (16 and up)? They changed it. The color has improved. The next generation will be interesting where they apply what they learned with the first camera and fix some weak points.

Did you use the new firmwares with the new color science (16 and up)? They changed it. The color has improved. The next generation will be interesting where they apply what they learned with the first camera and fix some weak points.

No idea I'm assuming they are debayering and converting using the latest builds , can't tell you anything about how it was shot , decent lenses though its nice and sharp.