New Sample Footage

[Philip Williams]

Originally Posted by Steve Mullen

<snip>
Moreover, there are no "mushy macro blocks" nor "noisey edges" in properly shot 24p or 30p. In fact the only sin in Brett's images is too much EE -- which he had control over. All the claimed V1U "problems" have appeared in video that was poorly shot. I've got many hours of correctly shot 24p video and so does DSE -- and we don't see problems.

If Bret had too much edge enhancement, why did some of the frame become sharp and noisy and some became soft and blocky? I had my wife watch Brett's video clip and she noticed how the front of the shed turned to mush. She actually said it "looks fake". Now trust me, for my wife to notice an artifact in a video its got to be pretty obvious. In Brett's sample, everything looked fine in interlaced mode. What in the world should he have done differently in Progressive mode?

Also, in Brett's video I didn't notice "noisy edges", entire portions of the picture became noisy - almost looked like a film grain filter.

Originally Posted by Steve Mullen

Shoot Sh*t and you should expect it will look like Sh*t.

Well the problem here is that sometimes perfectly fine interlaced shots turn cruddy when switched to progressive. I know you've got a zillion reasons why this is normal, good, acceptable or non-existent. But in every other camera I've seen the performance either stays the same or improves when switching to progressive. I think its reasonable to expect that a scene that looks good in interlaced will look at least as good in progressive mode.

Originally Posted by Steve Mullen

We are now in the seeing "faces in the clouds" world where reality no longer has any role. This is the internet at its worst.

Well at least we can agree on something.

As for the performance of progressive scan on the V1, perhaps we can simply agree to disagree?

[Tony Tremble]

Originally Posted by Alex Leith

Just out of interest, what did you get, Tony?

The only alternative, the Canon XH-A1. I am looking forward to just getting down and using the tool rather than being frustrated and writing about it.

Peace...

TT

[Tony Tremble]

Thank you Bob for taking the time to comprehensively and absolutely nail the issue down.

TT

[Chris Hurd]

Originally Posted by Stephen van Vuuren

I would love to see much more P and P vs I footage posted than what we have, especially some standard charts, controlled but varied lighting... I would be willing to host if DV Info had bandwidth issues as I think it would benefit the community.

Plenty of bandwidth here on DV Info Net that I'm happy to make available for this purpose. Interested parties who have such video that they're willing to share need only to contact me via email for the upload account info. Thanks in advance,

[Stephen van Vuuren]

Originally Posted by Chris Hurd

Plenty of bandwidth here on DV Info Net that I'm happy to make available for this purpose. Interested parties who have such video that they're willing to share need only to contact me via email for the upload account info. Thanks in advance,

Excellent - let's see some footage as there should be quite a few V1's in circulation now. Perhaps Tony can also shoot the same shots he did with his V1 as with his new A1 for comparison.

[Steve Mullen]

Originally Posted by Bob Grant

However what we're talking about here isn't a lack of resolution over the whole frame or a bit more than desirable noise. We're talking about an electronic artifact that isn't consistent accross the frame or over time. If it was as minor as the stuff in Steve's wonderful shots in the coffee shop no one will notice, heck I had to look pretty hard to find it.

It's when that becomes a major part of the whole image and there's nothing to mask it, no motion blur that pretty much masks the macroblocking in most HDV, no excuse that "nah it's not noise, it's grain mate"

Bob, IF what you decribed actually happened when shooting P then I would agree that there was a serious problem. But it doesn't happen! I've got hours of 24p shot from inside temples, inside houses, and in the blinding sun. Not once does what you describe appear in my video. Only a slight softness that you can see in in A B C, but can't see in 0 1 2 3. Which I assumed was the price to be paid for using "5" rather than "7." Now I think it is inherent in P mode.

Of course, I'm shooting with a very very early Sony prototype. Very likely the same one DSE used to shoot his stunning video for Sony -- before the NYC announcment -- that thousands have now seen. Unless we are both lying to you all, what you describe doesn't happen with prototype cameras.

I don't even want to think about what this might imply. I hate the idea that saying this will lead to even more speculation. Which is why I'm not going to continue in this discussion. I'll wait for a Sony to say something.

But, I'm sure in heck going to try to hold onto this baby as long as I can.

[Philip Williams]

In a thread on another forum its been indicated that an engineer with an american V1U has confirmed the "oil paint" effect in 24P and 30P. Hopefully we can expect more information by way of an article and some pictures within a few weeks.

Its not mentioned who the engineer is, but if its who I think it is, we should get a totally thorough and unbiased opinion on the issue.

I'm sure we'll hear more about this as the information presents itself.

[Bob Grant]

Steve,
sounds like I've had that V1U in my hands briefly too, well travelled camera!

Look I don't think anyone's telling fibs, I mean that most sincerely. I think we all need to take a deep breath and realise that IF there is a problem it's a damn complex one. As I've said before my analysis is that you could very easily shoot with this camera for the rest of your life and not feel there's any issue at all. I still think for the most part it produces some of the best looking images of any camera in it's price range and it's got enough image controls to keep any test pilot happy for years to come, I mean what image controls are missing on the dang thing, not many, it even lets you store setups onto flash memory, can't do that with even a Z1!

Now you mentioed the footage shot by DSE, I have to admit when I first saw it after other had pointed it out I could see the odd problem but a) it was a prototype and b) you had to look pretty hard to see the problem, sure didn't put me off one bit.

And then I saw your first posting of those coffee shop images and I was blown away, they looked great. Well OK, there was the odd bit that looked like CA but heck for the money no one could complain, every HDV camera has these issues. It's not the camera it's the glass mostly and you'd spend a lot more than what the whole camera costs just to get a lens that doesn't have issues.

But now we're seeing what were minor issues affecting small parts of the frame affecting large parts of the frame. No I don't think it's something that's crept into the production models that wasn't in the prototypes. There's a small trace of it in one of your coffee shot frame grabs. Well I think that's what it is and no I don't mean the wood grain. Take a look at the attached image I've circled an areas where I think the same thing is happening. In this case it's not a problem, no one would notice it without looking real hard and they shouldn't normally be doing that. But when this effect hits larger parts of the frame it's then that it cannot be ignored.

Now maybe there's a way to stop that happening, maybe I'm looking too hard, I don't have enough experience to really say for sure. All I can say is it looks wrong to me.

Now I'll admit in my most recent test I did something really dumb, I forgot to turn off the ND1 filter after the camera had told me to switch it on. Don't remember seeing it telling me to turn it off, stupid of me to not realise I should have anyway. One things for certain, causing the camera to wind the gain up sure bought the smear problem out in a high contrast scene. Again this shows two things. a) I'm a lousy cameraman b) Sony could make the indicator a bit more In Your Face c) the problem seems to relate to noise levels. Or perhaps part of the problem is people other than me not turning the ND filter off? And for what it's worth I was shooting 50i.

[Ken Ross]

Bob, I looked carefully at your circled area and I honestly have to say (even though you said the same thing), if we have to look THAT hard at an image to see an issue, then perhaps we should paint the picture instead (no paint-effects jokes please). I don't think there's another camera that could hold up to the scrutiny that this camera has.

Heck, I've looked at the Canon images and frankly I couldn't live with the terrible color seperation that seems to occur in some images on the right side of the frame. Just like the Sony, it doesn't show up in all clips, but IMO when it does, it's far worse than the issue with the Sony. It's particularly bad since this issue effects both the 'i' and the 'p' issue, it's inescapable. I've seen a number of clips that show this effect and it almost looks like a color CRT TV with terrible convergence issues. I myself could not live with that because it's such an 'in your face' problem. Frankly I'm utterly amazed that more Canon owners don't complain about that. My tiny little Canon HV10 HDV cam doesn't display anything like that problem.

So yes, none of these cams are free of issues, I guess you "pick your poison".

[Piotr Wozniacki]

Originally Posted by Ken Ross

Heck, I've looked at the Canon images and frankly I couldn't live with the terrible color seperation that seems to occur in some images on the right side of the frame. Just like the Sony, it doesn't show up in all clips, but IMO when it does, it's far worse than the issue with the Sony. It's particularly bad since this issue effects both the 'i' and the 'p' issue, it's inescapable. I've seen a number of clips that show this effect and it almost looks like a color CRT TV with terrible convergence issues.

Just my $0.02: you're right about the color fringing issue with the Canon, but IMHO it IS escapable - it only shows at the zoom and aperture extremities AND contrasty scenes all at the same time; you may easily avoid it (and BTW, just take a look into the "Purple outline on background objects" thread - I never saw a color separation that bad on the A1).

That said, I agree that the issue in the area circled by Bob is a minor one.

[Todd Giglio]

I have a V1U and (so far) I haven't noticed any of the problems mentioned. I have seen the 'purple fringe' in high contrast footage, but I have yet to see the 'oil paint' effect. I have to mention that I'm also using the M2 35mm adapter.

It has been mentioned that the oil paint effect happens in certain shots and not others. Could someone give me a shot list where they find these problems to be obvious and I'll try to shoot it. Trust me, if there is a problem, I'd like to see it before I shoot my feature.

I've been using the Z1U and FX1 for over two years and have looked at A LOT of foootage (and just to mention, I've never had a drop out on any of my camers; 3 of them).

I want to have faith in the V1U, and if the footage I've shot so far is an indication, I'm happy. Granted, I haven't really done any post production work on these clips, so I don't know how they would handle under color correction or extensive editing.

So... if there is anyone who would like me to test the camera under specific situations (and I don't mean just good lighting, I mean they way that you think the camera shows these flaws) let me know. Give me a shot list and I'll see what I can do.

Todd

[Michael Phillips]

Todd, I am not sure whether you are 24P or 25P but one of the most annoying issues I have with progressive 25P is on shiny edges.
Take a clip with a reflective edge in it as interlaced then take the same as progressive.
The interlaced has no problems whereas the progressive along the edges has a crawling effect and you can repeat it every time. You apply the cinema effect to the progressive, the crawl diminishes a small amount but you start to notice the darker parts of the image begin to loose detail (probably I could learn to live with that), but the uncertaintity of the image in regard to edges is not acceptable as when you take a scene you need a level of quality that is consistant.
I now just take interlaced to be on the safe side as progressive is too flakey to be relied on unless you have the conditions you can guarantee will not produce unwanted results.
I do realize that progressive is meant to produce a softer image to get the look you are after but it has to be consistant across the image.
Michael.

[Michael Phillips]

Could some of those who have observed problems with progressive try taking some shots with just manual settings.
I have done some of the shots I have had problems with and using a higher f stop or higher shutter speeds and fine tune it with a bit of gain, the crawling went and the image was sharper. I don't have the gear to get a detail look at the images and appraise them more carefully, but I would like someone else to check this out with better gear.
Michael

[Todd Giglio]

Michael,

I only use manual settings (especially since I'm using the M2) and I haven't noticed any problems either. I can't imagine that the camera would add these flaws if operated only in auto, but you never know...

Todd

[Steve Mullen]

Originally Posted by Bob Grant

Steve,
sounds like I've had that V1U in my hands briefly too, well travelled camera!

I don't think it's something that's crept into the production models that wasn't in the prototypes.

Bob, I just realised that I've had two V1Us. One came from DSE. And one came from DVexpo -- which might or might not be the first one I had.

I think it's obvious that Sony would not have given me the V1U twice if they thought there was anything wrong. In particular, they would not have asked me to shoot my Asia trip in 24p if they thought there was anything wrong with P mode. Chris will verify that I have been very hard on Sony -- and in public. They had to know I would be brutally honest if I found problems.

So we must assume that the video they and DSE were seeing didn't have significant issues. So let's call this Firmware A. And, I'm going to further assume (A) was what the Sony engineers thought was the OPTIMAL tuning given the way the camera works.

The version that went to the UK, which was the version Simon got, was horrible. Let's call this Firmware B.

Sony revised it for the production V1E's Let's call this Firmware C.

There may, or may not be any difference between the protype (A) and the shipping V1U. But let's call the shipping firmware, D. The difference bwtween (C) and (D) is the difference that has caused "many" V1E complaints and only a "few" V1U complaints.

All agree that P is softer than I.


------------------------

Why is the V1 P-mode softer than I?

A) 1080-rows become 1080-lines in a frame every 1/60th second. For P -- these 1080-lines are used in a frame that lasts 1/30th second. For I -- 540-lines are used for the first field. In the second field, a NEW set of 540-lines are used for the second field. Over 1/30th second -- a total of 1080-lines are presented. However, because the second field comes from a second capture, I mode will have more vertical information over 1/30th second.

B) 960-columns are "interpolated" to about 1440-pixels each time the CCD is read and processed by the EIP. For P -- these 1440--pixels are used in a frame that lasts 1/30th second. For I -- these 1440-pixels are used for the first field. In the second field, a NEW set of 1440 pixels are used. Thus, over 1/30th second -- TWO sets of 1440 pixels are presented. Therefore, when presented, a frame of I has more horizontal information than does a frame of progressive.

C) Thus as I've repeatedly said, given Sony's design, P will be softer than I.

------------------

My feedback from the UK is that Sony Japan has not given-up on 25p which is likely why 25P was restored.