Tony,

would it be a solution to shoot in 50i then deinterlace to make it progressive? How would that compare to 25p shots? Or am talking something stupid here?

Tony, it seems like when the blade of glass is not vertically straight-up got the most obvious "interlaced effect", was it very windy when you took the shoot ?

I am asking this bcs I realise that not only those few blade of glass in front of the swam got those effect, if you look closely to any area of the green, you can see those minor horizontal lines as well, not obvious bcs it is mostly green.

In this case I don't think it is the problem of the encoding.

Perhaps you can use the camcorder to take few pictures, see it gets the same issue or not.

Or perhaps your tape is not clean ? sound silly, sorry if i am stupid again.

Yes -- frame grabs add a bunch of variables. It's what folks see on a high-quality monitor that count.

An audience is going to the overall detail -- they are not looking for halos. This a nutty way to judge video. There's nothing wrong with your V1E. I wouldn't expect P to look exactly like I.

The fact is the stair-stepping is aliasing. And, yes it should be worse on diagonals. To eliminate it -- one would need to lower overall detail. Not a good choice!

Sony uses a setting of "5" for Cinema. I too want a detailed look so I wouldn't go lower than "5."

Well, it seems to me more like post-deinterlacing artefacts. I have no idea how on earth it appears in a progressively shot video, but when I feed and interlaced material to my 1920x1200 LCD through component and stop it, I'm able to spot a couple of edges with tiny spots of stairstepping like this; the same edges are stairstepped all-along in the original interlaced video, played back without any deinterlacing through DVI.

Tony, does it happen only in a recorded material, or can you spot it when the camera is connected live to a monitor as well, ie. before going to tape?

Steve, IMHO aliasing is more common with nearly horizontal, and not vertical, edges - just like the top TV set edge, heavily aliased in a picture posted in another thread on post-cure status of 25p...

I have to respectfully disagree on a couple of points.

I think the halos are a matter of personal preference, mine is to reduce them where possible. The aliased edges on fine detail is another matter. I simply don't understand why interlaced should produce a beautiful anti-aliased image and progressive produce an over sharpened image with aliased fine detail.

I personally would expect P to be near identical to I. Could you explain why you believe this not to be the case? Why does P footage have the aliasing while I footage does not?

Piotr

You have to be careful when monitoring on a 1920x1200 screen as that will induce stair steps if the screen changes the aspect ratio of the video. I don't have a way of monitoring live at native aspect ratio unfortunately so will not comment on live output.

Anyway, I'll do a bit more fiddling but am coming to the end of my time off. :(

Cheers

TT

Yes, they shouldn't be the same, but shouldn't P look better then I ? In term of clean & sharp resolution ?

Tony, my monitor has a capability to maintain the correct aspect ratio, so I'm always displaying at 1920x1080. So far, I never spotted stairstepping other than those post-deinterlace artefacts (should be called jagged edges, really).

In a word no!

Why?

Have you seen the difference between progressive and interlaced footage on a V1e?

TT

It's never made logical sense to me that a cam said to be natively progressive would use the same native progressive sensor in a model that only supported interlaced output, the FX7. I think we've reached the point through semantics where you can call something anything.

Add to it the 25p artefacts, and the original suspicion of the progressive being obtained through some internal deinterlacing returns:)

Poitr

Bad news I'm afraid. I have reshot my original test clip and the oil paint effect is still there along with over edgy lines. My original test clips didn't have enough high frequency detail to clearly show the effect hence my misplaced optimism. In a direct comparison between 25P and 50i the 25P material looks terrible. 50i never falls apart no matter what is filmed.

I am sorry for posting duff information before I had chance to fully test out the camera. I am hoping Sony just forgot to upgrade my camera by mistake.

Anyway my dealer is on the case....again. I am really impressed with their attitude now they are about to kick some a$$.

When you get your camera back could you do some tests please?

Sorry for posting duff info at the start of this thread...

TT

Tony,

I'm so sorry to hear this. Of course I'll do my tests as soon as I get the camera back, but I must admit I dislike all this more and more. Could you please drop me an e-mail with the information who your dealer is in the UK?

Thanks

Piotr

This was my original concern as well, why output as I if P is the native capture design.. but there are threads saying it is because HDV doesn't support P and Sony doesn't want to make incompatible format just like Canon did it 25f in XH-A1. I don't know this is the truth or not but any claim can have it's own very valid reason...

Yes, such "internal deinterlacing" does look obvious in Tony's screengrab, almost everywhere if you look carefully to the green field in the picture.

Tony, perhaps my initial suspicion is right, V1 is an overpriced FX7 with few more menu option plus a 90% good Progressive output (at least for PAL we live with). But it's 50i is really superb. I don't do progressive output so I won't have so much concern about it but if I haven't bought A1, I will stick to FX7 bcs no worth to pay couple hundreds pounds just for few more menu option.

Just my silly stupid opinion.

Hi friend, you have nothing to sorry for, I think Sony has to start listen to it's loyal customers like you and many of us.

And what you said might be correct too that they forgot to fix your camcorder before returning to you. I have similar experience dealing with Sony service center in Malaysia, my camcorder was like a ball being kick around few times before they really fixed the problem which is a month after few iterations.