Testing the 50Hz (PAL) XL-H1

[Steven-Marc Couchouron]

I've been testing the european version of the XL-H1 for a few days now and my results are fairly consistent with what has been mentionned in other threads by Shannon, Nick and Pete. Not only is it a superb piece of equipment, but it renders a very high quality image which truly seems to suffer less than the Sony from artifacting when the camera is moving. I am not going to list everything it has going for it since this has been done before. I'll just add that the stabilizer is much smoother than the one on the Sony.

I don't know if this is something other owners have experienced, but in full automatic mode, the XL-H1 I am testing has a tendency to overexpose by at least a stop. Not really a problem since I can use the AE shift, but still worth mentionning. Has anyone else experienced this ?

Another thing that I am not yet convinced with, is the "progressive" mode. Please bear in mind that this is the european model and that this mode is 25F only. On moving shots (not still resolution charts) I clearly 'see' a loss of resolution with 25F compared to 50i. Also, the judder seems to be more disturbing than what I'd been used to with the DVX100 - on some shots, but not all. However, I still need to do some further testing to really confirm these first impressions. I will also try to upload some comparative footage I've shot so that you can give me your own impressions. Just for the record, the Canon representative who lent me the camera specifically referred to the 25F mode as not being true progressive, and also saying that I'd be better off shooting 50i even for filmout (and let the lab do the work). Of course I'd rather do my own testing and base my opinion on the latter.

One thing I can say is that after having had it laying around for a few days, I am very jealous of those of you who actually 'own' it ! And this camera definitely has something about that makes all the heads turn when you're out shooting with it. If you want to stay inconspicuous, this is probably not the camera for you !

[Ron Pfister]

Thanx, Steven-Marc! I'm looking forward to hearing more of your impressions, as I live in PAL-territory as well.

Joyeux Noël!

[Karen Nakamura]

I'm in Japan right now shooting a short documentary film on a community of people with psychiatric disabilities in northern Japan. I had just gotten my Xl-H1 before leaving.

The XL-H1 definitely overexposes 1 stop or so depending on the situation. I think it's because of its center-weighted metering. But especially with ordinary room light, I'm dialing it back about 0.5 - 1.0 EV. This gives much richer colors. I wish it had a live histogram so I could see what I'm doing since it's a new camera to me and I haven't had a chance to output anything to a HD monitor for testing.

Still, the camera is absolutely amazing.

Karen Nakamura
http://www.photoethnography.com/blog

[Chris Hurd]

Hi Karen, I really like your site. Thanks for posting here, it's great to have you with us! Welcome aboard,

[Lauri Kettunen]

Steven, is the colors of the EVF neutral? The dark and midtones of mine seems to be biased towards magenta. (The view in the panel of Console program is neutral but in the EVF it's not.)

I also took some test shots and they are available at my pages: www.luontovideo.net -> Special

[Steven-Marc Couchouron]

Well, from my tests, it's not just the colors in the EVF that have a magenta tint, the colors from the footage also have it ! It's slight but it's there. And Karen is right, the metering in full auto mode isn't very good. As soon as you have some contrast in the image, it will overexpose and you'll end up with blown out highlights. Which means you almost constantly have to correct the exposure. Better to go manual.

I've done some more testing this week-end and am now going through it.

Want to see the HDV codec break up ? Go here:
http://www.dvforever.com/upload/CanonXLH1/IMG_0141.JPG
Of course this is the worst of situations, with the camera being still, but everything else in the picture moving.

A bit of smear, but not too extreme (this is the worst I would get under normal shooting conditions):
http://www.dvforever.com/upload/CanonXLH1/IMG_0147.JPG

Contrast range:
http://www.dvforever.com/upload/CanonXLH1/IMG_0123.JPG

These are highly compressed Jpegs, the actual footage is obviously cleaner.

[Ron Pfister]

Thanks for these images, Steven-Marc! Nice and tricky ones!! What were your image parameters during the recording of these clips (gamma, pedestal, knee, ...)?

TIA for any further info! Much appreciated!!

[Lauri Kettunen]

Originally Posted by Steven-Marc Couchouron

it's not just the colors in the EVF that have a magenta tint, the colors from the footage also have it !

Yes, I discovered the same today. Once I made more tests, I noticed the magenta tint also on the tape. (Interestingly enough, the very first footages from Kaku seemed to have the same property of magenta tint. )For the time being I managed to remove the magenta bias by adjusting slightly the values of the color matrices.

Speaking about these matrices, who knows what the entries are precisely? (A.J., are you there) My guess is that if one has a vector i=[r,g,b]^T of input RBG values and o=[R,G,B]^T of output RGB values, then o = Li holds, where the linear operator L is a matrix whose off-diagonal entries are RG, RB, GR, etc. But, when Canon writes RG, does the first character R refer to the input or output value?

It's surprising that Canon took the trouble to install such a menu of custom presets in the camcorder -as it did also in case of XL2- but then they still fail to specify, i.e., define properly what these custom presets really mean. For quite understandable reasons the definitions are not printed in the manual, but they could have established the proper technical specifications on some some web page. An unformal charcaterization is not a definition. Instead, such an attempt to unformally discribe the idea suffers in my view for ambiguity.

[Steve Rosen]

There are many such problems with the manual. I hope someone, maybe from this forum, publishes a book or pamphlet that describes in detail what all the settings mean in the real world.

I am a fairly experienced cinematographer but some of the options force me to do a lot of testing to understand them... not necessarily a bad thing, but when you're hot to shoot useable footage, who wants to sit around doing tests?...

There are many hints, ideas and sollutions posted in this forum.. wouldn't it be great to be able to access them on the run, refer to them from a pamphlet conveniently stashed in your bag.. get writing you guys...

[Karen Nakamura]

I would also buy an expanded XL-H1 manual or DVD, especially one that focuses on the arcane custom settings. It seems that there's a lot of black magic involved there. For an experienced DP, doing a short DVD on this topic might only take a weekend, and if 100 of us bought it at $30, that's an easy $2000 or so in profit...

[Robert Niemann]

Lauri, I have a questions regarding your "GREAT TIT" clip: Was the camera set to autofocus? It seems to pump.

[Lauri Kettunen]

Originally Posted by Robert Niemann

Lauri, I have a questions regarding your "GREAT TIT" clip: Was the camera set to autofocus? It seems to pump.

No autofocus, for that's not even available with EF-lenses. But, probably, since I had set the cache for 5 seconds in Conosole program, and the image is not shown in real time (and not even every frame) in the preview window, I may have turned the focus ring while the footage was recorded in the cache. Another explanation is, since the footage was taken about 7-8 meters away from the bird in very low light, the focal plane was also very very short. So, the tit may have also swung back and forth around the focal plane.