XM1-Look Preset?

[Mark Fry]

This started a couple of weeks ago when some friends sent me a "greatest hits" DVD, containing some of their favourite railway shots (that they had taken themselves) from the past 8 years or so. That sent me digging through other discs of my own shots and from other friends. One thing I concluded was just how much I like the look of the Canon XM1 (PAL version of the GL1). My sample consisted of shots from Canon XM1, XM2 and XH-A1, Sony VX1000, VX2100 and FX-7, and various smaller Panasonic and Canon cameras, often taken on the same day and in similar locations.

The most pleasing "look", to me, is the XM1. I'm reasonably sure that it's not down to one individual camera. My sample includes footage from three different examples, and they are much more like each other than they are like the other models in my sample. Nor is it down to post-processing. My friends and I aren't really into that, except where really difficult lighting conditions have made it necessary. Even then, we just tend to increase the contrast a little.

I know that the XM1 is not neutral, having a slightly warm (orange?) cast, but that's part of its charm, especially on a cold or grey day. I suppose it's like preferring Kodachrome 64 to Ektachrome or Fuji (which I do).

I'm going to do a few experiments, to see if I can come up with a preset for the XH-A1 that gives the feel of the XM1. But I'm a novice at this sort of thing, and I don't have professional test cards, lights, monitors, scopes, etc. I do have an NLE (Avid Liquid) with colour scopes built in, but I can see that it could be a rather long-winded processes, so all advice is very welcome, especially if it cuts down the time required.

My plan of campaign goes something like this:
Method:
- Print a simple test pattern (red, green, blue, magenta, cyan, yellow, white, black bars) on the best quality colour printer I can find.
- Put the test card on an easel, somewhere shady outside, with a clock in one of the corners (at the bottom of the black and/or white bars?).
- Set up both the XM1 and the XH-A1 side-by-side, white-balance them manually (off the same card), and record the test-card on both simultaneously;
- Load the footage from the two cameras into the NLE and compare them on the vector-scope.

Test 1:
- load some of the Presets posted on this site into the XH-A1, and compare them with the XM1. (Any suggesteions for good candidates?) Which is closest? Which least-similar? How does the appearance relate to the parameter settings and to the plots on the vector-scope? Is one of them close enough to act a starting-point for my own "XM1LOOK"?

Test 2:
- starting from the factory default, gradually adjust one of the XH-A1 colour-gain parameters, recording 10 seconds or so at each setting.
- reset back to default and do the same with each colour-gain and colour-matrix parameter in turn, to illustrate what each one does by itself;

I know that Test 2 only addresses the colour part of the equation. I'll have to think about a different method for black, gamma, detail, contrast and so on.

Has anybody done something similar already with the XM1/GL1? I've searched the "XH Custom Presets Download Library" thread for "XM1" and "GL1" but found nothing. It would be great to have another reference-point.

Has anybody written anything that describes how to adjust the colour-gain and colour-matrix parameters to make a particular effect? In particular, how do the matrix parameters work? I would like to read a bit more theory (so long as it's fairly practical!) before doing the experiments!

Any practical advice from those who've been here before me?

[Kees van Duijvenbode]

I recognize your story. I used a XM1 for 3 years myself and allways was very happy with the results. So when I bought my A1 I had hopes that I could match them and could continue to use my XM1 as a second cam. But after the first comparison of the footage of both cams I gave up. The difference between the two is, in my opinion, very hard if not at all configurable by means of a preset. It's not only a matter of colour differences but also a matter of sharpness, resolution, the count of pixels etc. Even in post I couldn't match them enough to my likings. At least not in the time frame I gave myself. That's why I sold it and bought a HV30 in return.

If I had lots of time to try it again I think I would start with comparing A1 footage with no preset at all and XM1 footage in a good NLE. That make's it a lot easier to do a side-by-side comparison. That footage should not only contain all important colours but also blacks, shade, bright sky, clouds etc. Preferable static shots. Then try to match it by changing the A1 footage with settings that could also be made in a preset. And make good notes. When you manage to match the pictures try to translate the changes you made with your NLE into analogue "preset changes". Make the resulting preset your starting point for fine tuning.

[Mark Fry]

I've begun my experiments, written up my efforts so far and posted them at http://www.steamage.co.uk/ColourTest/, and I'd be very interested in any helpful comments you may care to make. I'll add to it from time to time as I make progress - or otherwise!

Thanks in advance.