optical stabilization on XLH1 vs XL2 and XHa1
 

Reviewing some recent footage shot from a plane using an XLH1 it seems noticeably more stable than similar footage I took a couple of years back with an XL2 and also way better than footage taken that same day with an XHa1.

Is the optical stabilization on the XLH1 that different from the XL2 and XHa1?

depends on what lens you use.
some XL lenses don't have stabilizers.

Let's assume we are talking about the 20x lens since it is the only official XL-H1 series lens that even has OIS. In my opinion the OIS in this lens is better than the XH-A1 OIS. Not that the A1 is bad by any means, but the H1 just seems to give you much more stabile shots in similar circumstances. I don't expect you to notice any significant difference in average shooting situations but once things get a little more shaky or wobbly the H1 seems to pull ahead.

This is just my opinion of course....the form factor and weight of the H1 probably only further add to it's stability too.

i find that compared to the XL2, they are nigh on identical.. it wouldnt surprise me ifs theyre running the same algorythms.. (YES its all by numbers.. DVX100 and HVX are also by numbers.. speaking to a Pana tech, turns out that optical stabilisers are mathematically controlled to shift a centrepoint of the lens elements to comensate for motion.
Its not an actual physical piece of of glass which floats, umm how do i explain this... the OIS mechanism is the actual element which reads the inital motion. This is then sent to the cpu to look for corrections based on the lens position versus the XY axis of what teh ccd is is seeing.
From there, its THIS same piece which is then controlled by the cpu which tells it to shift its position to compensate this original shifting of axis... hard to explain.. but its all maths.. )
As youve used an XL, youll notice that the more you zoom in, the floatier it gets.. to a point of actually feeling rather soft. At full zoom, a slight shift should be easy to do, but it takes practice to perfect this with hand/shoulder control, as its easy to overshoot your frame when the stabilser floats this much...

the XL series lenses are "floaty" compared to the "solidness" of the A1's and the best way to describe them would be to consider shock absorbers on a car.
You can have soft shocks (XLh1) which "float" the car into stability, and the XL is very much like this.. to a point of sometimes overcompensating the motion. The A1 is pretty much the same at full wide, however as it has a shorter lens barrel, it "tightens" up as u zoom in. So that "floaty" feeling starts to dissipate.
I dont know why its not the same, but it i didnt design it, but this is the opposite for the XL cams, as they work better and get "floatier" as you zoom..

I cant explain it without mapping out the lens elements in relation to the barrel length in relation to the position of the actual OIS mechanism basied on the position of the lens elements themselvs at any given zoom ratio

I wish I could find the thread but very early on after the announcement of the A1/G1, Chris Hurd made a comment regarding the OIS of both cameras. He mentioned that one is a "vari-angled prism" style and the other uses something different (I can't remember off the top of me head). He went on to say while they are different systems. both will give almost identical results and that he doubted that anyone would really be able to tell the difference.

It sounds like we possibly stumbled into an area where there is some difference. If I have a chance to search the forums to find his original thread I'll post a link.

Also, having read the detailed description above I have to agree. The A1 is nowhere near as "floaty" as the H1 when zoomed all the way in. Having said that, the A1 is also less stable. Again, this is my opinion. Sometimes the H1 is a little "too agressive" and I can see the stabilizer trying to correct an image.....but overall it is great when you have to zoom a ton and are not on sticks.

Peace!

Definitely yes, re: XH A1. Sort of yes, re: the XL2.

Canon uses two separate methods of OIS for their camcorders: L-Shift and VAP.

VAP (vari-angle prism) technology has been around since 1992. Physically it's a bigger mechanism so it's used only in bigger lenses such as the XL series. The stock 20x lenses included with the XL2 and the XL H1 both use VAP for their OIS. This technology has improved over time with each new generation of lens. There's a year of development between the 20x IS lens for the XL2 and the 20x IS HD lens for the XL H1 (they are not the same lenses despite their similarities) so any difference that you can detect in OIS performance between these two lenses is most likely a result of developmental improvement (processing speed, response time, etc.) in VAP technology.

L-Shift (lens shift) technology has been around since 1998. Like VAP it is an optical image stabilization process using optical, mechanical and electrical components, however physically it is a much smaller unit than VAP and therefore can be used in a broader range of camcorders including the older standard definition Optura series and today's HV10 and HV20. It's also used in the XH series as well. L-Shift is very effective but VAP is still considered the flagship of Canon's OIS technology. Any difference you can detect in OIS performance between an XL H1 and an XH camcorder is most likely a result of the difference between the L-Shift and VAP technologies.

For a brief overview of both OIS types, see http://www.canon.ca/press/pr_content...1&id=155&arc=1 -- hope this helps,

Thanks Chris. I knew you'd come to the rescue with the answers and references. I learned some new things too! Personally I'm glad I still have my H1 for the VAP. Nice to know I am not imagining a difference between the A1 and the H1 also.

Peace!

thanks Chris et al

Yes I've definitely noticed that for me the XLH1 is significantly more stable in everyday use - but I put that down to the weight of the camera and better support.

But using the 3 cameras in an airplane supported by a cinesaddle those issues are largely factored out - and the differences are still there.