Harmonium – film or file?

Roy has already reviewed this film and I think he is gives a good sense of what makes it interesting viewing. However, he did not comment on the quality of the screening so maybe he was more fortunate than me. The film seemed interesting but I was constantly distracted by the poor quality of the image. To give examples; there were frequent long shots of a single character in mid-screen and mid-distance and the figure was fuzzy; there were also two-shots where the character farthest from the camera, but active in the frame, was fuzzy. Overall there was a lack of definition and, especially in the interiors, there was a lack of contrast. It seems likely that the film relied to a degree on natural light but even in this case I still thought the definition poor.

I saw the film in Cubby Broccoli at the National Media Museum (now renamed Science Media Museum and programmed by Picturehouse). I checked with Picturehouse and they stated that the film was screened from a 2K DCP.

So I then contacted the distributor, Eureka. They sent the following:

” The film has screened in multiple cinemas across the country and this is the first time that this has been brought to our attention. I apologise if your enjoyment of the film was spoilt in any way, and appreciate your feedback. I’ll pass your feedback on both to the cinema and our production team, because the film did screen from DCP at that venue – that is the only format that was made available to them.”

I followed up this reply with an enquiry about what was the source material for the DCP. This has not received a response. I have had other examples of this lack of response from distributors. I suspect it means they are not happy with the question.

This is where the ‘film or file’ in the title comes in. I was fortunate enough to attend the George Eastman Museum Nitrate Weekend; everything on 35mm. At a discussion I noted that archivists were distinguishing between:

‘film – i.e. acetate or nitrate relying on halide silver grain’.

‘file – digital relying on pixels’.

Even in the 2K digital cinema packages I am sure there is a difference between these two. However, it seems it is technically simple to up load digital video (DVD or Blu-Ray) onto a DCP. A projectionist I asked advised that once this is done there is not an obvious difference in the technical information on a DCP. But there is clearly a difference in quality. DCPs have technical specifications, detailed on a helpful page on Wikipedia (https://en.wikipedia.org/wiki/Digital_Cinema_Package). Likewise for Blu-ray [https://en.wikipedia.org/wiki/Blu-ray]. The comparison put simply is between something like 25 gigabytes for the video disc [or only five for DVD] and averaging a 100 gigabytes for DCPs. I gather DVDs do vary: some are lower than a 100; higher quality ones can get close to 300, and the latter take far longer to load on the server.

35mm prints vary: this was apparent at the George Eastman Nitrate Picture Show. And it has long been a point of comment by critics. One of the arguments for digital is that there is a common standard of projection. Using source materials below the specifications for DCPs clearly subverts this. Critics do not help as whilst they frequently comment on the quality of the style and technique they much less frequently comment on the quality of the actual screenings. [Roy does quite often do this]. Two friends did not remember noting a lack of quality when they saw the film but another friend remembered that he thought the image was not very good.

I tend to avoid reviews before a film because they so frequently reveal plot, character and even dialogue. And I avoid many trailers because they frequently offer a skeleton of the entire plot. But now I find I am researching films online beforehand to try and ascertain what sort of quality file is likely. I am also compiling a list of known suspects.


  1. Roy Stafford

    You may want to read this piece in ScreenDaily yesterday: http://www.screendaily.com/features/why-new-technology-could-be-hurting-the-film-business/5119128.article
    It sets out some of the issues and raises important points. I confess that I hadn’t realised that SMPTE was a different standard to DCI. Projection from DCP is certainly not foolproof and the article explains some of the background to why it may remain problematic for some time to come.
    I am surprised that you had problems with Eureka.


  2. keith1942

    Very helpful article Roy. It would seem that the industry is suffering from the same self-induced problems that plagued the transition to sound and to wide screen. This is, of course, how capitalism works: competition. But in all three cases cinemas were on the receiving end of not fully developed and inconsistently applied technologies. One thing that affects me as a ‘silent film’ fan is that the original specifications did not cater for frame rates below 24 per second.
    The other aspect, related to competition, is the drive to lower costs. My feeling about ‘Harmonium’ was that this was the problem: but it may well be that the factors set out in the article are the reasons.
    Roy obviously thinks Eureka are good at responding to queries. I have only limited experience with them. What I do find is that it is extremely difficult to get detailed information on film releases, especially when there is a process of film to digital file and, sometimes, back to film again.


    • Roy Stafford

      I don’t have any experience of contacting distributors about the quality of prints. I was simply referring to Eureka’s good reputation for the quality of their DVDs which include the Masters of Cinema Blu-rays and DVDS. I think Eureka only moved into theatrical distribution as a means of promoting their DVDs (I remember Tokyo Sonata as an early example in 2010). I thought you might support the label Keith since they have included instructions on how to watch their films in the correct aspect ratios – and sometimes offered two versions of a film title if it was originally released in different ratios, e.g. Shane (1953)). I would expect them to produce the best quality digital master for their Blu-ray release and that this would also inform the DCP. I’ll certainly look carefully at the next Eureka-released DCP.


  3. keith1942

    I wanted to add another example which is ‘The Lost City of Z’ (2016). I saw this in Pictureville in what was supposed to be a 4K DCP. There were ‘sploggy’ bits of colour early on and then in the World War I sequence there was ‘digital breakup’, blocks of pixels in the lower frame. I reported this to the projectionists who were going to take it up with the Distributor Studio Canal. I have not heard what, if anything, came back.
    However, I hear from a couple of other sources that they saw the self-same fault at other exhibitors, so this seems to not just be a one-off problem.


  4. keith1942

    Regarding Roy’s comment about Eureka and their Blu-rays/DVDs. In the case of ‘Harmonium’ I am commenting on a theatrical screening. I know that some distributors use the master for their DVD or Blu-ray as source for a DCP. It seems very difficult to check this. However, the quality of the image at the screening I attended made me suspect that this was an example.
    People who buy digital video are entitled to get a quality product, but so are people who buy tickets at cinemas. And the quality is not the same. A Blu-ray is about 25 gigabytes; the average for DCPs should be 100 gigabytes.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.