Press Release: Panasonic unveils HMC150 pricing and ship date

[Ethan Cooper]

Originally Posted by Brian Standing

Why is pixel shifting desirable?

Like far too many things in life, I'd imagine it's an economic trade off. Three 900ish x 600ish chips must cost a good deal less than three 1920ish x 1080ish ones.

[Darren Shroeger]

Originally Posted by Chris Hurd

I'm not aware of any AVCHD-branded camcorder that has the ability to record SD... anyone?

Although an SD capable camera was already mentioned, the AVCHD Information website has specs for the format here:

http://www.avchd-info.org/format/index.html

Interesting to observe that the 1080/720 signals all look to record with AC3-compressed 5.1 Dolby Digital audio, while the 480/576 signals get 7.1 Linear PCM. Am I reading this chart right?

[Robert M Wright]

I'm a little surprised 480/60p isn't part of the spec.

[Darren Shroeger]

If I ever need to get SD say for a quick-turn event I'd just haul my DSR-11 deck along and roll DVcam, or maybe even haul along the trusted Panny DMR-ES20 DVD recorder

Another question: Does the HMC150 output live video to all ports at once: HDMI, mini-D component, and composite?

My setup would benefit from this because I have a 12" focus check LCD monitor that runs off of HDMI, a Sony PHM-14M8U HD CRT client viewing monitor that takes HD component, and then the DSR-11 to record SD from the composite. So I could use all three outputs simultaneously in some shooting situations.

[Geoffrey Robson]

There's been some discussion about NTSC/PAL upgrades on the HPX-170, but I've not heard anything about this for the HMC-150. Does anyone have any information on this? Also, I imagine the "upgrade" is just a matter of activating some "hidden" functionality that's already on the camera, or is it actually a matter of physically adding an ASIC or some circuitry? If I'm right, then surely it's possible the procedure will leak onto the Internet, enabling anyone to upgrade their camera?

[Robert M Wright]

It will be interesting to see how the image quality of the HMC150 compares to the HPX170 and HVX200A. My guess is, in most cases, it will be noticeably better. 1920x1080 4:2:0 offers almost as much color information as 1280x1080 4:2:2, and almost twice as much luma information. 24Mbps AVC really should even hold up to motion pretty well (if the codec implementation isn't a dud). Once side by side footage starts getting compared extensively, sales of HPX170s and HVX200As may plummet. It seems to me that Panasonic may have made a mistake by not adding AVC-Intra recording capability to the HPX170 and HVX200A.

[Bob Diaz]

See a review of the HMC-150 LIVE:

Videomaker to review Panasonic HMC-150 Live on 8/19/08

August 11th, 2008 by jburkhart
Panasonic has given Videomaker the scoop on the HMC-150, and we thought it would be selfish to keep it to ourselves. So we’re going to be doing a live review of the first HMC-150 in the USA on Tuesday August 19th, at 11am pacific / 2pm eastern.

We’re taking it out for a test drive this week, and will report our impressions live at http://www.videomaker.com/live.

There will be footage from the HMC-150 to download, and the chat room will be active, so if you have a question about the HMC-150’s functions or features, join the live show and we’ll try to answer it for you right there.

P.S. If you can’t make the live show, the review will be looping on our channel afterwords at: videomaker.com/live

WOW, I can't wait for that....

Bob Diaz

[Bob Diaz]

I've converted the .pdf file form the Panasonic Government sales rep. into a form everyone could read here. The files might take a bit to load, but say a lot about the camera...

http://i46.photobucket.com/albums/f103/KQ6WQ/Page-1.png

http://i46.photobucket.com/albums/f103/KQ6WQ/Page-2.png

http://i46.photobucket.com/albums/f103/KQ6WQ/Page-3.png

http://i46.photobucket.com/albums/f103/KQ6WQ/Page-4.png

http://i46.photobucket.com/albums/f103/KQ6WQ/Page-5.png

http://i46.photobucket.com/albums/f103/KQ6WQ/Page-6.png

Of interest:

> Pre Record, 3 seconds!!!
> The 3 User Buttons have 11 choices
> PH (21 Mbps/24 max) & HA (17 Mbps) modes use Class 4 Cards
> HG (13 Mbps) & HE (6 Mbps) use Class 2

Bob Diaz

[Tim Polster]

Thanks for posting this Bob!

This camera has really caught my attention as a replacement for my PD-170/VX-2000 cameras.

[Alessandro Stefani]

Nobody know the european distribution date?

[David Heath]

Originally Posted by Brian Standing

It seems intuitively obvious to me that a chip with a pixel resolution that matches the resolution of the image you are trying to create should have an easier time of it than one that has to jump through some electronic hoops and interpolation to do so. It seems that if you want a 1920x1080 image, having a 1920x1080 chip will make your life much easier than having a 960x540 chip. Is that not true?

Simply, the answer to the question you put is "yes". But think of the implications of other than just the number of pixels, and the effect on resolution. Go from 960x540 to 1920x1080 and you must either make each individual photosite a quarter the size OR make the chip four times the area - use 2/3" chips rather than 1/3".

For sub-$10,000 cameras it should be obvious that compromises are going to have to be made - it's only a question of which is the best compromise, and this is where it starts to get hazy. Panasonic have decided to stick with 1/3" chips, keep fairly large pixels, and use pixel shifting to get as much out of them as possible. Does it improve the camera performance, relative to the same system with no pixel-shift? Yes. Is it as good as using chips of four times the area with four times as many pixels? Emphatically not.

The question isn't whether pixel-shifting works (it does), but HOW WELL it works.

Most discussion about it tends to revolve around resolution, but it's worth thinking about aliasing. That occurs when detail finer than the resolving power of the sensor gets picked up, and will appear as spurious coarse patterning. Unfortunately, pixel-shift techniques rely on detail finer than the native photosite dimensions to work, so a camera with effective resolution enhancement via pixel shift is likely to have higher aliasing levels.

Many cameras use it in the horizontal sense - Panasonic are unusual in using it horizontally and vertically. One unwelcome effect of this is reduce the sharpness of diagonal edges to enhance horizontal and vertical edges! It robs Peter to pay Paul. If you're playing the numbers game, it looks very good on test charts (where the resolution wedges are normally horizontal and vertical), but shows up on a zone plate.

Sonys approach is go for 1/2" chips in the EX series, and 1920x1080 resolution, the larger chip size meaning the photosites can still be reasonably large, and the real cleverness is in keeping the optical system size still to that of an average 1/3" prosumer camera.

[Brian Standing]

Originally Posted by David Heath

Sonys approach is go for 1/2" chips in the EX series, and 1920x1080 resolution, the larger chip size meaning the photosites can still be reasonably large, and the real cleverness is in keeping the optical system size still to that of an average 1/3" prosumer camera.

Hmmmmm... interesting... I'm getting (a bit) closer to understanding all this. Thanks for helping to enlighten me. And I suppose JVC's approach was to go with lower resolution (1280x720), but keep the chips at 1/3"?

[Bob Diaz]

... Sonys approach is go for 1/2" chips in the EX series, and 1920x1080 resolution, the larger chip size meaning the photosites can still be reasonably large, and the real cleverness is in keeping the optical system size still to that of an average 1/3" prosumer camera.

Just to add...

On the 1/3" Sony V7, the image sensor only has 1MP of sensing elements rather than the full 2MP. On the 1/4" Sony V1, the image sensor also has only has 1MP of sensing elements rather than the full 2MP.

The Canon A1 & H1 have roughly 1.5MP, BUT the low light performance suffers as a result.

The big issue is NOT what a paper specification says, but how does the overall image look? There are other parameters that impact the quality of the image, like noise, dynamic range, color saturation, contrast, ...


For some reason, I can't upload image files and the files are too big for my Photobucket account. I may have to crop the images from the HMC-150 in order to show what it looks like....


Bob Diaz

[Chris Hurd]

Bob, try this: Controls > Networking > Pictures & Albums

http://www.dvinfo.net/conf/usercp.php

As in http://www.dvinfo.net/conf/members/c...rd-albums.html

[David Heath]

Originally Posted by Bob Diaz

On the 1/3" Sony V7, the image sensor only has 1MP of sensing elements rather than the full 2MP. On the 1/4" Sony V1, the image sensor also has only has 1MP of sensing elements rather than the full 2MP.

The Canon A1 & H1 have roughly 1.5MP, BUT the low light performance suffers as a result.

Quite right, but 1MP is still twice that of the HVX200s 0.5MP, and there seems to be something of a concensus developing that for 1/3" chips 1MP may be a sweet spot in the compromise - more means that individual sites become too small, less just isn't felt adequate, pixel shifting or not. And that 1MP is being offered in a variety of ways - 960x1080 (Z1), 1280x720 (JVC), and the 45degree diamond arrangement of the V1, Z7.

It took me a while to properly understand how the latter worked, but the more I found out, the more impressed I became - it enables a 1MP chip to have square pixels, yield equal resolution horizontally and vertically (and roughly equivalent to a conventional chip of 1440x710), but still be easy to process in a 1920x1080 matrix. No other 1/3" chip design can claim all of those factors.

The big issue is NOT what a paper specification says, but how does the overall image look?

To a point, but sometimes paper specs can give a clue to important issues that just looking at overall images doesn't! Aliasing would be a good example. Aliases may not show up very much in first generation images, but can really mess up final compression. And generally the less pixels in the sensor, the more likely aliasing is likely to be a problem.

Similarly, and regarding general theory behind pixel-shift techniques, it's also worth emphasising how it may work very well for some images, but hardly at all for others. It's at it's best for subjects of low or zero saturation - a white/black edge can affect the green CCD or the red/blue CCDs, hence (via pixel shift) the system resolution can easily be seen to be more than any individual sensor. But what about a green/black edge? It can't have any effect on the red/blue CCDs, hence pixel shift techniques don't contribute, and the resolution is purely that of the green sensor. That may not matter very much with most real world images, but what about chromakey? In that case the resolution of highly saturated images becomes very important, and pixel shift may let you down.

Frankly though, as far as the HMC-150 goes, I think overall image quality may be less relevant than other factors in making a purchasing choice, and in particular computer processing power required to post produce. I can't help thinking that the HMC-150 may give comparable quality to HDV, but at lower data rates, with the penalty being more computer power needed. So to compare with the Z7 - 25Mbs MPEG2 to Compact Flash may be as economic as a slightly lower bitrate H264 to SD, whilst still giving the solid state advantages....... whilst the MPEG2 is far easier to post produce!

And the Z7 has the option of tape AS WELL as solid state.........