|
|||||||||
|
Thread Tools | Search this Thread |
December 1st, 2005, 05:02 PM | #16 |
Trustee
Join Date: Sep 2005
Location: Rio de Janeiro
Posts: 1,240
|
Look, this is ridiculous. I came here for help, for knowledge. I want to make my research as deep as possible, and that is why I looked at toons of footage, asked a toons of people with very different backgrounds, and it's why I came here. Some people here are saying that I came here and started saying a bunch of unfundamented bad things about the Fx1...but instead of fundamenting you own answer to WHY were my comments wrong, you simply started acting like I insulted your wife or something. If somebody misinterpreted it as an insult or offensive, tough luck. I've already said that was not what I meant and I'm not going to repeat it again. This is not personal, guys! I WANT to embrace Hd. I really do. So, instead of keep saying "boohoo, you're ugly and dumb!", say "I used this equipment and I can say for a fact that what you have heard and read is wrong...and here are the facts...". If you want to focus on the boohooing me part, please open another thread saying how much you'd like to nail me to a cross and stone me to death. Whoever is matures enough to remain on this thread, please, help me up here. I have an important investment to make and just want to be sure of how capable this tool really is. The ag-dvx100 and the Fx1 are more less the same price and I need to decide which would be the best solution for me NOW...in 4 years there will be better cameras and new threads like this for guys like me to be beaten to a pulp. I know that because of the "in your face" look there's a lot of people adopting the FX1/Z1 for documentary, but that's not the look I'm going for, since I subsist on music videos and commercials (with the ocasional narrative project). Anyone working with music videos/adds of using the FX1/Z1? I'm really curious to see/hear about the results? I've seen some FX1 footage with a 35mm adapter and it's the most film like I've ever seen.
|
December 1st, 2005, 09:00 PM | #17 |
Major Player
Join Date: Mar 2005
Location: Ontario, Canada
Posts: 916
|
Rafael, your original question was "How does SD DV look on an HDTV?" Perhaps your question should have been "HD or SD?" Given that you've provided commercials and music videos as your staple, here's my thinking on the subject. I recently dug up a paper copy (nothing on the net) entitled "Understanding Camera Resolution", from Broadcast Engineering, August 1999. I've cut and pasted the text of my comments on that paper below. Film is of course far beyond the digital standards described there..but HD is as close as we can get. A recent grad from the local college film course suggested that working with 16mm film was not far off from current HD production costs. Maybe you should investigate working with the real thing!
If you are aspiring to that rather elusive (and completely beaten to death if you ask me) film look, and you have all the other elements in place, then an HD cam with a 35mm adapter is the next best thing to 16mm. Based on the article below, that is obvious. For every HD problem out there, I've read corresponding fixes and work-arounds. Early adopters of any technology will always pay more, and figure out the bugs for late adopters like me :-) Even if you never master any HD projects, you will have better footage than native SD cams by downsampling. If you are buying a new camera, HD to me is a given. If I was loaded (not drunk, just rich) I'd have already pre-ordered an HVX200. My advice is to wait and see what the HVX will bring to the table. By the way, by responding to your post, I in no way am insinuating that I am in any way mature, responsible, or even know what I'm talking about :-O "Understanding Camera Resolution",Broadcast Engineering, August 1999 The article describes interlaced transmission and some of the problems associated with it, including interline flicker, line crawl, and the visible field line structure (the visible field line structure masks perception of detail in our visual systems within an octave of the field line structure). These all have an impact on perceived visual resolution beyond what a static resolution chart would suggest. In other words, it?s worse than it looks. The tests involved a high resolution scan of fine pixel images which were pre-filtered with a cut-off of 1.3 times the Nyquist limit of a coarse sampling pattern. They then post filtered with a cut-off at the Nyquist limit for display. The original oversampled images were filtered with different two-dimensional cut off frequencies. In a forced choice comparison the sampled and filtered images were compared. The subjects were asked, ?Which image looks sharper?? When the choices were 50-50, they compared resolution of sampled, vs un-sampled images. The spot size of the un-sampled images was .7 that of the sampled. What this all boils down to is that it took twice as many pixels in the reference image, to produce the resolution of a properly filtered sample image. From those comparisons, the conclusion was that progressive images can have a resolution higher than the Nyquist limit. With a sharp cut-off filter, a progressive image has a 2 fold increase in perceived vertical resolution over an interlaced image. EIA1596 charts are included in the article to illustrate the differences. A few interlace filtering interim solutions are suggested to address the fact that our visual systems have a slow response to detail and that motion detection is at low resolution. In one technique, the top octave of detail is updated at 30fps (fast enough for detail perception) and the rest of the field line structure updated at 60fps to maintain good motion rendition. The article goes on to reinforce that a full progressive system can produce a perceived increase in resolution by a factor of 2. Keeping in mind that this work was done in 1998, the same forced choice experiment is done to compare the various HDTV specs. A chart details the perceived sharpness of various displays equated in equivalent pixel number: 480i - . 25 Megapixel 480p - .5 Mp 720p - 1.45Mp 1080i -1.7Mp 1080p - 3.2Mp These numbers are explained with a few observations. At normal viewing distance, the limiting resolution on a viewer?s retina is about 22 cycles per degree. That equates to 32 million pixels in our field of view. At normal viewing distance, NTSC TV subtends .5 percent of our field of view, 1080 line TV, 720 projection, and 35mm projected subtend 5%, 1080 line progressive subtend about 11%, and IMAX about 67%. The article concludes that using a progressive camera, with 1080 line progressive display of interlaced transmission, properly processed, can provide the best perceived sharpness of the options up for review. Nothing surprising there..the military and medical communities have been working on this since around 1978. Nyquist limit: http://www-camil.music.uiuc.edu/cla...udio/audio.html |
December 2nd, 2005, 02:43 AM | #18 |
Trustee
Join Date: Sep 2005
Location: Rio de Janeiro
Posts: 1,240
|
Thank you very much, Dennis. That was a very insightful article/comment. Like you said, the much anticipated HVX200 is bound to open new doors, since it'll both record in HD and progressive. Conserning these numbers:
1. 480i - . 25 Megapixel 2. 480p - .5 Mp 3. 720p - 1.45Mp 4. 1080i -1.7Mp 5. 1080p - 3.2Mp Number 3 would correspond to the ag-dvx100 series and number 4 would correspond to the FX1, right? The numbers are pretty close. By the way, I'm a Pal shooter, so a little bit more resolution can be added to the equasion. Once more, thank you for this very helpfull article. Cheers, Rafa |
December 2nd, 2005, 04:41 AM | #19 |
Major Player
Join Date: Nov 2004
Location: Portsmouth, UK
Posts: 611
|
Rafael,
The DVX corresponds to 1 or 2. All those numbers refer to the vertical dimensions of the raster (ie how many pixels top to bottom) the DVX would have 720 pixels across. ALso, I think the 720p and 480p refer to 50/60p rather than 50/60i (actually 25/30i) Anyway I have some limited expereince with the DVX but extensive experience with the Z1 and it is an excellent DV and HDV camera, if 1080i is your thing and the 24p/25p issue isn't important to you. It has very high level of professional controls, the audio is very good, and the input options extremely flexible. If you're going for HDV forget the FX1 and get the Z1. If you were to use it as a PAL DV camera, in cineframe25 mode it would be VERY close to producing a full 25p SD image. Its image quality in SD is slightly inferior in some regards to the DVX but a lot of this could be down to a matter of taste, Sonys tend to have a rather "harsher" more clinical look compared to Panasonics and Canons, (in my opinion) but actually I quite like that look, and of course it can be subtly tweaked in camera and in post. The DVX is one of the best 4:3 SD DV cameras I've ever used (not having had had any experience of the XL2) but as an SD camera, the Z1 comes close, and then if you factor in that it does HDV (so far I've had one drop out, on aan old, re-used multiple time tape and the artifacts aren't that bad unless you REALLY push it) and native 16:9 I think it's worth the difference. There is no crappy audio on the Z1, the compression is hardly noticable except barely on extremely sound critical areas, i.e. I recorded a solo acoustic guitar in a recording studio with a Senn 416 and think I heard some very light compression swirling on a couple of chords, but I had to listen really hard. Bear in mind that the DVX DOES NOT have a native 16:9 chip, so in fact any widescreen material you should will be upressed from 720x432 (though shooting squeezed camera may give better results than cropping/uprezzing in post, maybe a uprezzing-plug-in writing DVX user can chip in here) - or of course you could use an anamorphic adaptor. |
December 2nd, 2005, 05:37 AM | #20 |
Trustee
Join Date: Sep 2005
Location: Rio de Janeiro
Posts: 1,240
|
Thanks, Dylan. That was very helpful! Did you ever shoot music videos, commercials, narrative projects with it? I'm very curious about the results. Most of the stuff I've seen out of the Fx1 was documentary or live TV.
|
December 2nd, 2005, 08:37 AM | #21 | ||
Major Player
Join Date: Mar 2005
Location: Dallas, Texas
Posts: 234
|
Quote:
The FX1 would be #4 (in 1080i mode) The HVX200 would be #3 (in 720p mode) or #5 (in 1080p/24 mode). Quote:
|
||
December 2nd, 2005, 09:46 AM | #22 | |
Major Player
Join Date: Nov 2004
Location: Portsmouth, UK
Posts: 611
|
Quote:
There are a number of pop videos being shot with the FX1/Z1 and sometimes with 35mm lens adaptors. Do a search on thos forum, there's a few posted here. |
|
| ||||||
|
|