The link is: First Look: Sony HVR-V1 24p HDV Camera (http://www.dv.com/features/features_item.jhtml?category=Archive&articleId=194400711) by Adam Wilt.

Free registration with DV.com is required to view it.

The active area consists of 1080 staggered rows of 960 photosites each. It's as if the even rows were pixel-shifted compared to the odd rows. Each row, however, is sampled 1920 times: each photosite is sampled by itself and a sample is taken in between photosites, combining information from the two photosites on either side and the ones immediately above and below. In essence, each scan line comprises two sets of interleaved sample types: 960 pin-sharp samples and 960 more diffuse samples averaging the four surrounding photosites. Sony claims that this gives the V1 full 1920 x 1080 resolution.

The question I'd like answered, is how do they state in the owners manual (for FX7) that the max sensor is 1440x810 (for still pics) 1.16mp (less for video) yet they are sampling video on the same chip at 1080 rows???

The other question mark is about the rows being scanned 1920 times, meaning each pixel is scanned twice. There would be no need for that double scan unless the dynamic state of the pixels was expected to change after the first scan. Put another way, this is very much like interlaced scanning, except the first scan being sharp, the second fuzzy as Adam explains.

I am not convinced this sensor is 960x1080, promo info notwithstanding. It simply can't be if the sensor specs for the FX7 are germaine to the V1U.

TI am not convinced this sensor is 960x1080, promo info notwithstanding. It simply can't be if the sensor specs for the FX7 are germaine to the V1U.

You are assuming Adam is exlaining it correctly. My explanation in HDV@Work is different and I think is more accurate. TheFX7 and V1 are the same and are 960x1080.

I found the article very disappointing. Little more information than is already in the public domain. A "me to" article if there ever was one.

Res chart information could easily been posted with caveats that was a pre-production model etc etc.

One interesting point of note though was the reference to the Cineframe Matrix (3d LUT) unfortunately it looks like it's not able to be modified.

TT

There isn't very much deep diving a reviewer can do with a pre-production MT sample. Any journalist worth his or her salt knows this going in. That's why articles of this kind that are published before the camera is released will never include such things as res charts, etc. That's Journalism 101 for any industry trade publication.

You are assuming Adam is exlaining it correctly. My explanation in HDV@Work is different and I think is more accurate. TheFX7 and V1 are the same and are 960x1080.

Agreed but what your explanation doesn't account for is what's stated in the specs in the FX7 manual, which is not pre-production. Steve, I'm not saying you're wrong, and I'm trying not to assume anything.

I thought Adam's article was excellent, as was Simon's and yours. But like the mystery in the beginning about the HVX200, this one's not adding up for me.

Tom

Do you have a link to the FX7 manual? I've seen several references to it but have not found it yet.

Cheers

TT

I don't know where that link was but I can email the pdf to you?

Chris totally correct. Adam wouldn't post numbers on a prototype nor would I. The only numbers I used are from DSE because I know he would test it correctly. And, then only because they matched what Sony was publically saying.

Also, I'm not sure my explanation is correct either! I'm leaving for Asia tonight and haven't had time to really get into Adam's story.

However, I am reasonably sure he is wrong about the rez differences between P and I. I found NO difference -- which is what led me to a different explanation of HOW it works.

I think Sony's explanation leaves a lot OUT while my explanation, which tries to describe every detail, has to be taken as "speculation" until confirmed by Sony.

By the way, this week's HDV@Work is on the V1's lattitude.

http://digitalcontentproducer.com/newsletters/

The one before is on 3ClearVid and the one before that is on CMOS. Please read them in order unless you are a CMOS guru. :)

Imager Device
4.5mm (1/4 type) 3CMOS sensor
Recording Pixels (HDV/DV16:9 still recording):
Max. 1.20 Mega (1440 x 810) pixels – note 2
Gross: Approx. 1, 120,000 pixels
Effective (movie, 4:3);
778,000 pixels
Effective (movie, 16:9):
1,037,000 pixels
Effective (still, 4:3):
778,000 pixels
Effective (still, 16:9):
1,037,000 pixels

note 2:
The unique pixel array of Sony’s ClearVid Sensor and image processing system (new Enhanced Imaging Processor) allows for still image resolution equivalent to twice that of the image sensor’s effective pixel count.

1280 horizontal seems as likely as 960...

Thanks to Steve Mullen for the three articles on CMOS technology, and applications in the V1... Really sheads a light on things...

...allows for still image resolution equivalent to twice that of the image sensor’s effective pixel count.

If we accept that, then by extension that's 2.4mp for a resultant still image size of 2065H x 1162V. If the enhanced processing works the same way for video, you'd be picking up about one interpolated sample for every 2 real ones in both directions, horizontal and vertical, which makes more sense to me that Sony would try to do that than to have full resolution in one direction, and all the interpolation in the other.

Best Guess? 1280 x 810

1280 horizontal seems as likely as 960...
Sony has confirmed 960x1080.

Sony has confirmed 960x1080.
Pardon my ignorance in this matter, but can someone explain this to me?

The camera is 1440x1080, so... what's the 960x1080 mean exactly?

What is an "effective pixel"?

And if I bypass the compression and record straight via HDMI to my PC do I get 1920x1080 or do I get some other strange combination of numbers?

Like the Z1/FX1, the sensor size is indeed 960x1080, but thanks to pixel shift, non-square pixels and more, it goes to 1440. I'm still getting my head around the engineering of the V1, but from what I understand, it's a little different than the Z1/FX1, but still produces a great picture!

heath

...the sensor size is indeed 960x1080, but thanks to pixel shift, non-square pixels and more, it goes to 1440...

Yet I can get 1920 through HDMI? Man, every now and then I feel content that I am on top of the technical mumbo umbo that goes on with these cameras but then I realise how little I understand...

You and me both, my friend. The HVX goes from 960x540 to 960x720 and 1280x1080.

heath

I don't know where that link was but I can email the pdf to you?

That is very kind of you to offer.

I was really only interested to see if there is any mention of CINEFRAME MATRIX in the manual. I wonder if there is if you could paste the section if it exists?

---
I have just pre-ordered a V1e....Roll on Dec.

What really swung it for me was the latitude this camera has. Thanks to Steve Mullen for his excellent article on latitude in the HDV@work series.

Latitude for me is far more important than raw resolution and I hasten to say the FX7/V1 looks no slouch in that department either. I think the improvement in latitude that the FX7/V1 offers sets itself apart from the competition for me by moving away from the contrasty look of video. This is overwhelming more important to me than slightly less light sensitivity.

The camera may not be any good for those that want to shoot chain link fences in the dark but for all others this should a fine investment.

TT

One thing that still concenrs me is how good the mpeg2 encoder in the camera is. You can have detail up the wazoo and great latitude and such but if the encoder isn't very good then the video will have a lot of compression artifacts. The problem with detail is that it is harder to compress. I personally wasn't very impressed with the mpeg2 encoders in the older SONY HDV cameras. Canon seems to have one slick encoder in their cameras which is something that makes me very happy. With HDV I am more concerned about how well the image will compress in the camera.

I have heard nothing about how much better the new encoder is.

The video which Sony showed on a 40 foot screen at the press conference was really impressive. I shoot a lot of nature scenes with my Z1, and have always been a little disappointed in shots that have a lot of foliage detail. The V1 demo reel had some landscape shots which I thought were stunning, and I don't think my Z1 would have handled those scenes nearly as well. There was one shot of a field with tall grass blowing in the wind which really made an impression... You could see every blade of grass.

"Latitude for me is far more important than raw resolution and I hasten to say the FX7/V1 looks no slouch in that department either. I think the improvement in latitude that the FX7/V1 offers sets itself apart from the competition for me by moving away from the contrasty look of video. This is overwhelming more important to me than slightly less light sensitivity."

I tend to agree. I have mine on order from B&H. I am interested in low light sensitivity, and I think the progressive shooting with 1/30th or 1/24th shutter speed may help in some circumstances. I think people have been barking up the wrong tree for a few years thinking that other features (I won't mention as I don't want a flame war) give a cinema image quality. Resolution is fairly important, but most people watch their movies on a good ole 640x480 television these days (in the U.S.). It is the origination quality that makes the difference and people are fairly happy with the quality of DVD movies on a CRT television. The originating production quality and usually the use of film have been the deciding factor of quality in the past and now the new HD cameras with decent resolution and higher exposure latitude are closing the gap. I saw some film clips that were originated on a 2K camera and shown upsampled on a 4k projector and they were absolutely perfect. I dare say it was better than film!

I can't wait! :) The FX7 images really have me sold and the V1 should be even a bit better with more image controls like knee and black stretch.

I agree that the biggest IQ improvement still to be made is DR, but Adam Wilt's pre-review leads me to not get overly excited about this camera, yet:
"Dynamic range and noise on the prototype looked roughly comparable to those of the Z1, but I'm withholding detailed judgments until a shipping version of the camera is available for test."

If there was a solid 1-stop extension in its highlight range, that could make up for a stop or so loss in sensitivity, but something tells me we won't be getting that until the 1/3" version of this chip. -Still, ANY improvement in DR is welcomed.

If there was a solid 1-stop extension in its highlight range, that could make up for a stop or so loss in sensitivity, but something tells me we won't be getting that until the 1/3" version of this chip. -Still, ANY improvement in DR is welcomed.

Using the same "tricks" as we have used on the HD100, I am sure the latitude is very wide: KNEE = LOW and STRETCH = ON. But, it may not be wider than a Z1. But the CMOS image has less contrast than the CCD image so it seems wider -- more natural.

The CINE MATRIX has no "matrix" control other than Color and Hue. This will work fine for those of us who want to leave COLOR alterations to post where one can experiment. For those who lug a monitor into the field and feel safe making a comitment to a look during shooting, this is going to be a "missing feature."

My sense of the V1's size and cost means it is a "gun & run" HD camcorder so I'm not surprised that it has exactly the feature it has.