XF300's new competitor

[David Heath]

Originally Posted by Christopher Young

The added bonus with the 250 is that it has built in flash band compensation in its hardware / firmware. An absolute positive compared to the Canon or Sony cameras for that matter.

I disagree with that. Whilst the flash band compensation (all makes) may do a good job most of the time, it can get fooled occasionally - and then the result is really bad.

But in software you have the choice of approve it or not. If in hardware, that's it - the "correction" gets burnt in at time of shooting, end of story. For that reason, I'd say the Canon/Sony way of doing it is preferable overall.

After experiences with the 301/371, I'd be looking at the chipset and ghosting issues very carefully. The 371 turned out not to be a new chipset, but aggressive inter-frame noise reduction, which was how it appeared to be more sensitive than it actually was. Maybe the 250 chipset has improved matters, but it's certainly a point I'd want to check out very carefully. (On the 371 it was evident even without gain. Didn't really show too obviously on most pictures, but do something like chromakey and it fell apart.)

[Christopher Young]

Yes David, I here what you are saying re the flash banding issues.

I don't have too much experience with CMOS flash problems from a post point of view as ninety-nine percent of the material I work with is from HD CCD cameras and so have never had the problem of trying to correct such problems, have seen them for sure but thankfully never had to fix them.

Having used the earlier 300 Panasonic models I was aware of the smear artefacts which were quite evident at almost any gain position. This smearing is what I was looking for on the 250. Didn't notice any at the time. I was told that the 250 uses a new 3D adaptive noise reduction approach as opposed to the earlier models that used temporal recursive noise reduction algorithms. I guess it's wait and see to see what the factory finally delivers. As I said, it was only a short hands on with the camera so purely some quick subjective observations.

Chris Young
Sydney

[David Heath]

Originally Posted by Christopher Young

I was told that the 250 uses a new 3D adaptive noise reduction approach as opposed to the earlier models that used temporal recursive noise reduction algorithms. I guess it's wait and see to see what the factory finally delivers.

The trouble with any in-camera noise reduction is that whilst it may clean up pictures nicely at first sight, it is not the same as a camera with inherently less noise. Even the best algorithms "clean-up" the most subtle parts of the picture along with the noise. They may not be missed for most viewing - but their absence can be significant when post-processing is attempted. This was the case when trying chroma-keying with the 371 - the noise trails were barely visible on normal pictures - but a problem for the keying process.

It's also worth saying that "3D" in this context means the dimensions of width, height, - and time (frame by frame). In other words, a temporal algorithm. "Adaptive" in this context is generally understood to mean varying the noise reduction averaging to make allowance for motion - if you didn't, did it on a simple frame by frame basis, that's when you get ghost trails on moving items. But doing it in a more complex way - averaging pixels only where there is no motion - must mean no noise reduction where there IS motion. You get noise trails rather than actual "ghost" trails. Undoubtably better, but surely that's the same as the 371 did?

I'm still left feeling that electronic noise reduction may be best left to software in most cases, for the same reasoning as flash banding reduction.

[Galen Rath]

Do we know who is making the lens for this camera?