It has never been so difficult to decide on a camera purchase...

Easy choice - get the one with pretty colors! Ooh no wait... get the one with all the cool buttons and stuff to mess with. (laughs)

Actually Tim, even if you have tons of cash it's still difficult. You can go for the HPX3000 but no slomo, Varicam 2700 60P but only 720. Sony F23 and SR deck, I think that gets you 1080/60P but you've got a huge size and weight issue. Phantom HD's a great camera, and can work well for 25 fps work but it's far from a "normal" camera in operation. RED? Well plenty of issues there still.
Hmmmm.!

My assumption had been that although the imagers in the 2700 were each 1280x720, it was employing pixel shifting techniques to bring the luminance resolution beyond that of the 720p system?

Hence, an effective (luminance) resolution of the order of 1600x900. (Assuming both H & V is used.) May not be up to what 1920x1080 imagers may manage, but a front end performance ............
My apologies for misleading people, but I've just been informed that it appears the 2100 (so presumably the 2700?) DOES NOT employ any form of pixel shift to enhance the resolution above that of the native sensors.

Possibly even more surprising (in a camera in this price range) seems to be the absence of any optical low pass filter. The camera therefore exhibits higher than expected levels of aliasing.

...The camera therefore exhibits higher than expected levels of aliasing.

Is this fact, or assumption? Have you seen output from the camera to verify this?

Is this fact, or assumption? Have you seen output from the camera to verify this?
The best reference that has been forwarded to me is: http://www.bbc.co.uk/rd/pubs/whp/whp-pdf-files/WHP034_ADD24-Panasonic-HPX2100.pdf and in particular look at page 13, section 2.2.1, and the associated images.
"There are clear null zones, where the wanted lower frequencies beat with the unwanted alias products, caused by the presence of higher frequencies, at 720 vertically and 1280 pixels horizontally. At first sight, this is to be expected from a camera using sensors of 1280x720 dimensions. However, it actually reveals that no tricks have been played in offset positioning of the sensors, precision horizontal offset of green from red and blue is a common trick to extend the resolution, and this can be done vertically as well. The resulting performance is a little disappointing. It is also evident that there is no optical spatial filtering in this camera, to suppress frequencies higher than can be resolved."
And the images presented do seem to back up the researchers conclusions.

I'm normally extremely wary of reading too much into reviews seen on the internet. But that this has been done by the BBC R&D department and is officially posted on their site gives it an authority that few reviews can claim - it's good enough for me to consider it as "fact".

The lack of pixel shift techniques is one thing - that was, I freely admit, my own assumption - but I am extremely surprised at the absence of an optical lowpass filter in a camera at this price point. I wouldn't expect to find one in something like the Z1, HVX200 etc for both cost reasons, and also that the cheaper lenses are soft enough to tend to do their own limiting. The 2100 is likely to be used with much more expensive 2/3" lenses, far more likely to give severe aliasing without a good OLPF, and that seems to be what is being seen here.

If there's any lesson to to reading test charts like this, it's that they have little to do with real-world usage. If we all used test reports like this as the sole basis for buying any camera then most people would never purchase anything.

Every camera has it's weaknesses and I've seen charts like this from more expensive cameras with worse results.

At the end of the day it's what the output looks like that matters, not measurbating tech-specs.

At the end of the day it's what the output looks like that matters, not measurbating tech-specs.
Well yes, it is what the output looks like that matters, but from a broadcasters perspective - or anyone who shoots for broadcast - there's a lot more to it than just looking at the output of the camera or the edit suite. It's what it looks like in the viewers home that counts, and that's why I'd argue that test reports (at least scientifically done ones) do matter very much. For reasons far from arguing over just whether camera A looks a little bit sharper than B.

As example, I've posted this link and quote before (I suspect Digibeta is a typo for HDCAM): http://tvbeurope.com/pdfs/TVBE_downl...s&Analysis.pdf
NRK principal engineer Per Bohler was receiving calls from leading newspapers in Norway asking why the first HDTV pictures from Germany were so poor.

“I had to admit it was poor quality, and at first we couldn’t explain why. The EBU satellite feed was fine, giving us MPEG-2 422 profile at 24 Mbps. We recorded it to DigiBeta, and our transmission output looked good when it left us — but the viewers received disappointing pictures.

“It really astonished me that the pictures from the satellite looked so good, but collapsed so quickly when we compressed them for transmission. It seems that concatenation of different compressions from acquisition, to the EBU and on to us, meant all the headroom in the signal had been lost by the time it reached us, with nothing left for the last encoder to work on,” he said.
The main problem is the cascade of different codecs, different bitrates - all OK by themselves, but mixing in a cocktail to give unpredictable results where it most matters, in the quality of the picture at the viewers home.

And that's why worrying about aliasing is not "measurebating", because aliases can really screw up the final low bitrate compression at the very end, having been hardly noticeable through most of the chain. What's really nasty (from a compressors point of view!) is that they move in the opposite direction to moving objects they are associated with, and that can waste a lot of bandwidth.

It's also worth noting another bit from the BBC R&D analysis:
"Aliasing can cause difficulties in post-production operations where keying is used, since edges are not always where they are supposed to be."
I'd understood "measurebating" to mean picking over minor numeric spec differences. I don't call the absence of any optical low pass filtering in a camera in this price bracket minor, and neither did the source who alerted me to my earlier error.

I wouldn't expect to find an optical lpf on a camera in the price range of such as the Z1, HVX200, V1 etc, and as far as I know they don't have one - you get what you pay for. But the HPX500 seems to have an anti-aliasing filter, in spite of being much cheaper than the 2100. Don't you find that odd?