picture quality A1 vs FX1

6 replies [Last post]
bryan_g
Offline
Joined: Dec 19 2005

Hi,

Can anyone tell me if the A1 (1 CMOS) delivers a better (or equal) SD/HD picture than the FX1 (3 CCD) ? The reason why I'm asking this is because the A1 is considered a pro camera and the FX1 is called a consumer camera. I don't care about audio capabilities because I mainly do music video's (shooting and editing) and my clients deliver the audio on CD.

- Low light performance is very important to me.
- I'm willing to downconvert to SD if the final picture is better than when shooting native DV

Would it be better for me to just purchase a 3 CCD SD camera ? The DVX100, PD170 and even the panasonic GS400 produce exellent pictures, but lack true progressive 16:9. I've done some video's with a rented DSR 570WSP and though the quality is great I don't like shooting with this cam. My final output will be DV (for digibeta transfer) and DVD (for distribution) for the next couple of years, because HDTV and media are still "future talk" in europe.

Unicorn
Offline
Joined: Apr 12 1999

Not sure about the two cameras, but odds are you'll get a better picture on Digibeta by shooting HDV and downconverting to uncompressed SD than by shooting DV: downconverting from HDV will give you four times as many color samples as DV (twice the color detail that Digibeta can record, rather than half as much).

Also, if you want progressive SD output, throwing away one HDV field and scaling the other field (or putting black bars at the top and bottom) should be pretty close: one HDV field has 540 lines, one PAL frame has 576.

P4-3.06/2GB RAM/2500GB IDE/SATA. Avid Media Composer, Liquid Edition, Premiere 6, Lightwave, Vue 6, eyeon Fusion 5. DV and HDV editing/compositing.

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

I've done tests at DVC. Here's the numbers.

FX1/Z1 has 3 ccds, each 972x1080. The green is offset from red/blue by a half-pixelo (standard procedure in 3 sensor cameras) so delivers approximately 50% more horizontal resolution than the pixel-count suggests, i.e. 972x1.5=1458, the limiting resolution on HDV2 (1440). Vertically, it delivers 540 on each field by line summing in interlace, which subjectively appears to be about 50% more, i.e. 540x1.5=810. When set to film-mode it drops alternate fields so delivers only 540.

HC1/A1 has a single cmos sensor of 1920x1440 pixels, and I suspect it's a conventional Bayer pattern, so it delivers horizontal resolution of about 3/4 the pixel-count, 1920x3/4=1440. Resolution above 960 is confusable with aliasing, so it's not possible to give an exact number, but subjectively it delivers 1440, just. Vertically, it delivers 540 on each field in interlace, which subjectively appears to be about 50% more, i.e. 810. When set to film mode it apparently delivers 1080 although resolution about 540 is confusable with aliasing for the same reason.

So, subjectively, the HC1/A1 is sharper than the FX1/Z1, although some of that sharpness is confused with aliasing.

In other ways, the HC1 is very like the FX1, A1 very like Z1, in the range of tweaks available in the menus. From a purist's poiht of view, the FX1/Z1 is a good interlaced camera that does a reasonable job of film look, the HC1/A1 is a proscan camera that does a reasonable job of interlace. Overall, I prefer the HC1/A1 approach even though the sensor really ought to have more pixels because its lens is sharper than it can cope with (it produces strong aliases at 1080).

Hope that helps.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

Richard Payne
Offline
Joined: Sep 15 2000

In regards to the science - What he says ^

In real life I just filmed my daughters Christmas play on an HC1 and FX1 in HDV. The HC1 had a wide angle adaptor on it and was at maximum wide. I was using the FX1 for closeups.

Probably because the HC1 was wide, with a wide angle adaptor as well, it produced a much better exposure. I would imagine the gain was full on, but as I had about 1 minute to set it up I chickened out and left it all on auto. On the HC1 and A1e it decides when to use gain.

Even with 18db on the FX1 at 50th and wide open the pictures are 1-2 stops darker, and the noise is more evident. Of course I was using it about half way down the zoom range so I guess alot less light was getting in.

In reality the HC1 was a star in the gloomy conditions and the FX1 a bit disappointing, even though on paper this shouldn't be the case.

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Agreed, my preference would go for the HC1/A1 for consumer/prosumer stuff, FX1/Z1's too big anyway.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

bryan_g
Offline
Joined: Dec 19 2005

Thanks for the info. I've downloaded some HDV footage and converted to SD. Also found some great comparisons on the net. Downconverted HD is clearly a winner, if done properly.

I've seen the HC1 and if the A1 is just as small than it won't impress my clients. I think I'll be going for the FX1 or wait a little and go for the HVX.

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

A1 is identical to the HC1, but has more menu control, no flash, and the audio box on top. HC1 and A1 are aimed squarely at the consumer end of the market.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.