Two low-light frame grabs...

[Steve Crisdale]

Two JPEGs that might be of some interest...

http://www.pnc.com.au/~scris/hd_imag...105014100.JPEG
http://www.pnc.com.au/~scris/hd_imag...105014222.JPEG

If you can't already tell, the lighting in this venue was appalling to say the least.

Captured using VisionDTV DVB-t player software, from 19.3Mbit progressive program stream.

Program stream rendered in Vegas from 25Mbit FX-1e PAL m2t.

[Marc Martin]

what was the gain used on these 2 frames grabs?

[Steve Crisdale]

<<<-- Originally posted by Marc Martin : what was the gain used on these 2 frames grabs? -->>>

18Db.

[Anhar Miah]

I dont hardly see any noise in these grabs, except maybe the curtains but only very little.

You say 19.3Mb/s i thought HDV was 25Mb/s or are you using JVC HD10?

[Marc Martin]

There is some artifacts in the blue and red chanel (I think it is mostly MPEG2 artifacts), but it appears to be usable.

[Steve Crisdale]

<<<-- Originally posted by Marc Martin : There is some artifacts in the blue and red chanel (I think it is mostly MPEG2 artifacts), but it appears to be usable. -->>>

Which is why I decided not to make any comment about the quality of the stills.

They are JPEG'd straight from my Digital Broadcast stream player, which will account for some of the artifacting, but as noted: it does appear to be useable.

The reason I dropped the bit-rate from 25Mbit to 19.3Mbit (yes, I do have a HD10u) was to match the highest Bit-rate I've encountered for a High Definition Free To Air broadcast, and see how the result 'stood up'.

Given the comments about requirements for broadcast I've spotted flying around, I thought I'd see if it was possible to make something that could be put to air without any further transcoding.

I was also somewhat taken aback by Barry Greens still images from the FX-1 that were acquired under better lighting conditions, yet showed inferior results, and I felt the need to address what I saw as a somewhat tainted assessment.

[Davi Dortas]

Barry Green same guy who says FX1 downrezz footage look worst than DVX100.

[Steve Crisdale]

<<<-- Originally posted by Davi Dortas : Barry Green same guy who says FX1 downrezz footage look worst than DVX100. -->>>

Yes; that Barry Green.

I must say that after creating a DVD of the Airshow footage I took with the cam (the first video I shot with it!!!), I was amazed at the quality. It's obviously from video, but the detail retention made even commercial 'blockbuster' DVDs look soft and blurry by comparison.

After reading Barry's assessment vs the DVX100 I was left wondering how I'd managed to achieve a result that was so much superior to his.....

Some people require faith to survive the obvious I suppose. I'm glad that Barry has found his faith in the DVX.

[Barry Green]

The answer is simple... on our tests we didn't use any gain on any of the cameras. On your shots you used 18db.

The gain on the FX1 is pretty clean, even at +6 and +12 there's little noise, although it does start to get kind of noisy at +18. But in the comparison we did, there were just too many combinations to test, so we decided to use 0db on all of them.

And hey, regarding credibility, I'm not looking to gain or lose any, just reporting what I saw -- and making the footage available to anyone who wants to see. On an HDTV the FX1 blows away (to smithereens) the standard-def cameras. But on a DVD, it doesn't offer any advantage, picture-quality-wise; it looks about the same, excepting it's a couple of stops slower than the SD cameras.

If anyone has any footage anywhere that shows an FX1 or Z1 looking better, on a DVD on an SDTV, I'd sure like to see it. Testing should use comparable settings for edge enhancement/sharpness, of course (out of the box the FX1 is set to be quite a bit more "sharpened" than the other cameras, which may account for some of what you're seeing).

I guess we could always just put both cameras on max sharpness, to remove that as a variable...

[Khoi Pham]

That is not a fair way to test a camera Barry, the reason for manual adjustment is for the camera man to adjust the camera to its best under the situation, you made no judgement on the situation and left the camera at 0db, it could be way under exposed under the situation but could be perfectly exposed with the other camera, you criple the camera and then test it, how fair is that?, the camera would have perform better even if you leave it under auto, at least it would know to gain up some under the situation, you need to adjust each camera to its best under the situation then compare, if you don't have time to do that then it is not a good and fair test, no wonder why your dvd is not any better, because you encoded it from a crappy under exposed image.

[Steve Crisdale]

<<<-- Originally posted by Barry Green :

If anyone has any footage anywhere that shows an FX1 or Z1 looking better, on a DVD on an SDTV, I'd sure like to see it.

-->>>

On an SDTV......

Anyone considering purchasing a HDV camera to produce material they're going to watch on a SDTV, has got rocks in their head!!!

What about levelling the playing field properly by watching those comparison DVDs on a WS HDTV?

Those people - like myself - who've invested in any of the HDV camcorders, are aware that people with older SD equipment will be unable to see the full quality that HDV provides. And yes; we're also aware that DVDs are the only truly viable means of distributing the material shot to the vast majority of clients.

Despite the loss of quality in going to DVD, we'd also like to hope that those DVDs will be seen on a reasonable monitor/TV that's at least WS.

Of course, given that not everyone has WS or HDTV the most likely scenario is the 'lowest common denominator' one you have outlined Barry.....viewing on a 4:3 format (with inherent resolution reduction from letterboxing) with radically fewer scanlines than any basic HDTV.

Given that any DVD I make using my HDV material won't look any worse on a 4:3 SDTV, than that made using any of the SD camcorders you tested doesn't make me want to change my decision about purchasing the FX-1 or any other HDV camcorder for that matter.

Whether other people decide otherwise is up to them, but at least for the sake of helping people make those decisions with some assurance they do so knowledgeably, it's worthy of those advising to be as honest and demanding as we would expect from those advising us.

[Douglas Spotted Eagle]

Having shot an awful lot with the Z1, I too, don't quite know where Barry is coming up with his assessment, but I do know I respect what he has to say, even if I disagree w/him. I guess I'm the same way about Vegas as he is with the DVX100a.
I do have this to offer, I shot some images in Hawaii using a DVX 100a set to auto, and the Z1 set to auto. The weren't on a tripod, as I was in the fish market. (We shot other things too, but the fish are particularly telling) The third cam was a Sony 700 with a low end Fuji HD/ENG lens. No auto on that cam.
I've encoded 90 frames from all three cameras into SD MPEG 2, and dropped them both back onto the Vegas timeline, and extracted a still from both. Not a very scientific test, but in a short time on an airplane, it let me play a little. Nothing was done to the images whatsoever, other than the encode, and the placement on the Vegas timeline after the encode, then the PNG, as that's the only low-loss format that Vegas exports.
So, if you'd like to pass your own judgements, albeit that this isn't scientific because it's a fishy way to judge images....
http://www.vasst.com/HDV/Fishyinfo.htm
BTW, they're all pillarboxed, because that's what Vegas does in the widescreen mode when working with MPEG.

[Barry Green]

On an SDTV......

Anyone considering purchasing a HDV camera to produce material they're going to watch on a SDTV, has got rocks in their head!!!

Er... well, that's harsher than what I've been saying, but yeah, that's along the lines of what I've been saying. Actually, that's a lot harsher than what I've been saying -- I've been saying that once downrezzed, it produces video that's comparable to, but maybe a little softer than, the SD cameras.

What about levelling the playing field properly by watching those comparison DVDs on a WS HDTV?

What's not level about the playing field? I said that when viewed in HD mode on an HDTV, the FX1 blows away (to smithereens) the SD cameras. Completely. On an HDTV, the FX1 delivers footage that is FAR superior to the DVX and the XL2. How is that not a level playing field?

And yes; we're also aware that DVDs are the only truly viable means of distributing the material shot to the vast majority of clients.

Which is all I was saying -- for the immediate moment, for the near future distribution will be on standard-def, so it seems prudent to examine the cameras' performance on standard-def. On DVD. That's the only practical viable distribution medium.

Obviously if HD-DVD were a viable alternative today, the FX1 would have a H U G E advantage. A staggering advantage. But, by the time HD-DVD is available in a large percentage of homes, it's likely we'll be on our third or fourth generation of HDV (or other high-def) cameras. And the FX1 will likely not be at the head of that pack, just like the VX1000 is not competitive today against the DVX, XL2, and FX1.

I don't understand where the confusion is coming from. I've said repeatedly that on an HD television, there's no contest at all! If you intend to display your work on an HDTV, the FX1 and Z1 can do a great job -- they deliver video quality that is, in comparison to SD, jaw-dropping. If you need a high-def camera, the FX1 is a good choice and the Z1 is a GREAT choice. The question I'm raising is, are you SURE you need a high-def camera? Because there's not a lot you can do with it, and you do give up some options for no tangible benefit in the final product (assuming the final product is a DVD).

On a standard-def DVD, I haven't seen *any* improvement in quality from having shot on an FX1. Instead I see the loss of the options to have shot 24P, the loss of the 30P option, the loss of a couple of stops of sensitivity, and the loss of XLR audio. And the net result on SD is a picture that, at best, looks about as good as, but not better than, the best crop of SD cameras.

So my conclusion has been, if your work is going to be distributed on DVD, you're not really gaining anything by having shot it in high-def, and you are losing some things.

Why is that so hard to understand?

Despite the loss of quality in going to DVD, we'd also like to hope that those DVDs will be seen on a reasonable monitor/TV that's at least WS.

Yes, a better end delivery medium is in all our hopes. HDTV has only made it to, at best guess, maybe 7% of US homes. Widescreen is catching on faster than HDTV is, and the pace of adoption for 16:9 is accelerating. Probably 10% of the TV's in US homes are now 16:9, probably more like 50% of the TV's in British homes are 16:9. I viewed the DVD footage we shot on a 32" 1080i broadcast monitor, a 13" TV, a 10' LCD projector, and a computer screen, to get a wide variety of experiences before feeling that I had enough info to make a qualified judgement.

Given that any DVD I make using my HDV material won't look any worse on a 4:3 SDTV, than that made using any of the SD camcorders you tested doesn't make me want to change my decision about purchasing the FX-1 or any other HDV camcorder for that matter.

Well, yes, that right there has been my point all along. The FX1 delivers DVD's that look about the same as the other cameras do. So you have to determine what is most important to you: HD resolution, or 24P, 30P, XLR audio, and 2 stops of sensitivity. That choice is the choice you make, because currently no one camera gives you all of that. And on DVD, which is *the* reigning distribution medium, you can see (and your production can benefit from) the 24P framerate, the 30P framerate, the better audio, more image controls, and the better light sensitivity. But you cannot see the benefit of the HD resolution. That's reserved for either broadcast, private play off a computer, or some day years in the future when HD-DVD is a reality.

Is that position really unreasonable?

Whether other people decide otherwise is up to them, but at least for the sake of helping people make those decisions with some assurance they do so knowledgeably, it's worthy of those advising to be as honest and demanding as we would expect from those advising us.

I thought I had been, and continue to be. I agree with what you just wrote, 100%. And I've said repeatedly -- for HD distribution, the FX1 slays the SD camcorders where they stand. But HD distribution is not a reality today for most content creators, excepting some private corporate/tradeshow work and a few HD-only networks. Nor is it likely to be any sort of substantial reality for several years. So you have to determine for yourself if the "coolness" factor of shooting HD, or the "future-proofing" value, is worth the other sacrifices you have to make, because currently no one camera "does it all", so you will sacrifice something one way or the other. The choice is up to you as to which you prefer -- which is how a properly-operating marketplace should function.

So as I see it, the practical matter, the PRACTICAL matter is, what can you do with it today. TODAY. I set out to find out. And I reported what I saw. And I posted the results and the framegrabs too, so you could make your own decisions. And I invite anyone to disagree with me, as is their right and as is proper, 'cause I sure don't know it all. I just haven't seen any evidence to the contrary yet. That doesn't mean it won't arise, I just haven't seen it.

[Steve Crisdale]

Thanks Douglas!!

Your images speak louder than words.... more so as my browser didn't identify which image was from which cam (I did figure out the Sony 700 image was the last one).

There's just an extra sharpness to the Z1 image, that seems to confirm exactly what should be logically correct; i.e. that DVD level video created from HDV should retain more data than that created from SD sources.

Thank goodness HDV doesn't include an Aroma stream!! Fish markets can be 'on the nose'.... Hope you got there early!!!

[Barry Green]

Originally posted by Douglas Spotted Eagle : albeit that this isn't scientific because it's a fishy way to judge images....

Spot, those are beautiful shots. The Z1 definitely has stronger contrast. The difference with the 750 picture is huge.

Here's an idea: let's see if we can't set some time at CES to shoot something and come to some sort of conclusion. Is Chris going to be there? Let's take your Z1, Chris' XL2, and my DVX and set up a shot where we can A/B/C them on a monitor, and use the image controls in each camera to their best ability to see how close we can come to making each shot look like each other. (judging color, or "warmth", or even sharpness, is very difficult on these cameras because they offer so much variety). Once we eliminate those as variables, let's see how each compares in delivering any particular look, on DVD. That way each camera will be handled by its respective biggest advocate, should provide for an interesting shot.