Jon Fordham's HDR-FX1 Review

[Charles Papert]

Jon:

On my recent feature, we had the Sony 20" HD monitor at the engineering station, and two 8" SD monitors for A and B cam monitoring for the director and script super., etc. The SD monitors were being fed from an Evertz 9155 afterburner. One of the monitors was my personal 8045Q. Camera was an F900.

Every now and then I'd catch sight of the 8" monitors, which had usually been tampered with, blacks raised up into oblivion (script supers generally prefer it this way so that they can see details in the shadows--to be perfectly honest, of all the scripties I've worked with, there are very few that I would be inclined to consider their opinions on picture quality). I'd calibrate my 8045q monitor on the odd occasion if I needed to watch takes from those monitors. My impression from bouncing back and forth between this setup and the 20" was that the image was, as expected, night and day. The color rendition and contrast were pretty comparable, but a small SD image vs a large HD one, especially on a broadcast monitor--there was beyond "snap" as a differential! I felt that the image on the 8" could have easily been generated by a Digibeta, or really any decent 2/3" camera. Definitely I wouldn't have put money on a 1/3" DV camera.

As another example: last year I did a studio shoot with the SDX900, and there was a DVX100 floating around on set (the director's personal camera). At lunch I cabled the DVX to the 14" broadcast monitor and A/B'd it with the SDX, under the same lighting conditions. Some thought the image was amazingly similar. I could definitely see the difference, although considering the price gap, the DVX did very well.

So Jon, to clarify the missing "snap"--would you possibly agree that this was largely due to the size of the imager and associated processing, in other words the camera head itself, rather than the medium (DV vs HDV)? Since both the DVX and the FX1 were 1/3" cameras with similar lenses, wouldn't it be expected that when viewed in an unbiased SD environment, they would appear more similar than compared to an high end HD camera? And furthermore, would a downcoverted F900 to that same 8045 monitor look significantly different than a, say, 790 Digibeta?

If you do have the opportunity to test the output of the FX1 with the F900 on a broadcast HD monitor, please let us know what you think about that, it would be interesting.

In the meantime, I thought your assessment of "which camera to shoot for the DV (SD) format today: FX1 or DVX100" was a useful approach for most users, since the HDV delivery path is still in its infancy.

[Heath McKnight]

I can probably do an F900 vs. FX1 test; a company I freelance for shoots with the CineAlta and is considering an FX1. I can probably bring the camera out on a CineAlta shoot.

heath

[Ken Ross]

<<<-- Originally posted by Troy Lamont : Jon,


Anyway, agreed that the FX1 probably isn't up to the standards of the high end HD cameras but Sony has definitely taken a step in the right direction and also shamed JVC in the meantime.

Troy -->>>

Troy, I got my first opportunity today to play with the Sony FX1. WOW! I hooked it up to my calibrated 50" Fujitsu HD plasma and the picture was simply mind blowing. I had tried the JVC when it first came out and was disappointed with most aspects of its picture quality and wound up returning it. But the difference between the Sony and the JVC is like day and night. I'd swear that some of the test footage I shot (on a dreary day here in N.Y.) could have been broadcast on HDNet and I wouldn't have thought twice about it...it was THAT good. I honestly had never expected the "pop" of the picture. I think the picture was exceedingly close to the scenes as I saw them with my eye. There was no hype, no push, just the scene as it was. Colors dead on, light dark values very appropriate etc. Needless to say I am very very impressed.

My only disappointment with the unit thus far is the auto focus. It seems that it misses on occasion. I had a subject standing outdoors in the center of the frame and the camera chose a lawn chair in the background to focus on. My VX2000 doesn't do that. I'm not sure if there are any adjustments that can be made in the autofocus mode other than simply going manual.

[Steven White]

I must admit, I'm curious about two things after reading this review:

#1. We all know there is resolution loss in Cineframe24. I'm not surprised therefore that with about a 50% loss, and downconversion Cineframe24 looks like similar resolution to 24p miniDV. I'm curious how the FX1 and DVX would compare on 30p or 60i footage?

#2. The DVX has an advantage over the FX1 for stuff like water because it's not using the MPEG-2. How does the FX1 pure DV compare? Obviously few of us are going to use the DV feature, but it really hasn't gotten much evaluation.

To me, comparing the FX1 to an XL2 or DVX requires a comparison with the DV codec. Someone should do a test predending the FX1 is a DV cam, and see how well it performs in comparison.

-Steve

[Ken Ross]

<<<-- Originally posted by Jon Fordham: It amazes me that some people will find whatever they need to discount the findings of an experienced professional in an effort to support whatever idea or belief they desperately need to cling to.

Jon -->>>

Jon, I am truly sorry you took what I said so personally. It was not meant as an attack on you or your methodologies. My major point was that most people simply can not perceive the necessary detail on an 8" monitor, regardless of the monitor's resolution, to make meaningful evaluations of all the extra resolution that the FX1 has compared to a DV cam. A far better comparison can be made on a larger HD screen where the additional detail (or lack thereof) can be more easily discerned by most people.

I think you probably could have been clearer in terms of how the cameras were hooked up, what the monitor was capable of etc. I think this has caused some confusion not only on my part but others as I read their posts.

Again, I am not trying to defend any purchase (I haven't made one) and I have no pre-conceived ideas about the Sony. However, it is obviously logical to assume that an A/B of both the Panasonic and the Sony on larger screen HD monitors, would show the far greater resolution capabilities of the Sony compared to the DVX. I got my first opportunity to play with one today, and I've surely never seen any DV camera come close to the resolution and sharpness of the Sony. I had tried the JVC when it first came out, but was totally unimpressed.

So please, don't take my post personally.

[Barry Green]

#1. I'm curious how the FX1 and DVX would compare on 30p or 60i footage?

CineFrame 30 fares a little better than CF24, because the motion rendition is constant and the resolution is constant. You get about 575 lines of vertical res out of CF30 (and, presumably, out of CF25 as well). So CF30 vs. 30P, you'll have about the same resolution vertically, but of course the FX1 will have twice the horizontal pixels when displayed on an HDTV. I've compared CF30 vs. 30P from the other cameras on DVD, and the results are consistent with the results I've posted before.

Regarding 60i performance, well, it depends on what you display it on. Displaying on a DVD or straight on an SDTV they'll be similar (I have some footage of this) but displaying on an HDTV, the Sony spanks the other cameras thoroughly.

How does the FX1 pure DV compare? Obviously few of us are going to use the DV feature, but it really hasn't gotten much evaluation.

The FX1 as a straight DV camera looks pretty good, it's native 16:9 and the only DV camera with an actual 16x9-shaped chip. But it still is a couple stops slower than other DV cameras like the VX2100 or DVX, and a good stop slower than the XL2. It has more vertical smear than the other cameras as well. Its performance is probably on par with something like the PDX10.

[Ron Evans]

Stephen, my comparitive comments in my earlier post were both HDV ( water/panning etc) and about FX1 in DV mode which I will shoot in most of the time for the next few months. I bought the FX1 as a 3CCD DV camera with good controls and HDV as a bonus. To repeat. Since getting the FX1 early December I have used at 4 dance recitals in 4x3 DV mode 60i ( about 8 hours video) , about 30 mins in HDV shooting scenes when I first got the camera ( water in the harbour, leaves blowing off trees , then snow falling/blowing etc finally checking out features like programmed transitions etc) and used 3 tapes of family events over Christmas in HDV. All tapes standard Sony Premium DV. I rent DVX100 and AG DVC200 so have reasonable knowledge of how they perform in the theatre environment. At 0db the DVX100 is apparently more light sensitive but I and others feel that there is already gain applied on the DVX100 at the 0 db setting. For me at 6db, in a dark high contrast environment, the DVX100 is not usable because of increasing grain and colour noise, but I would use the output of the FX1 even at 18db if needed. To explain, where the DVX100 would be at say 0db and F3.4 ( about the usual settings in the theatre) the FX1 would need to be at 6 or 9db at F3.4 ( my quess since I don't have them next to one another to confirm, based purely on shooting video in the same theatre most of the time). I like keeping the iris at F3.4 to F5.6 to give the best depth of field on stage so that I don't need to continually focus. So for usable video the FX1 may be more light sensitive ( having the extra USABLE gain settings) I have no experience with the VX2000 so can't comment. I won't rent the DVX100 again until May time frame but will do a more detailed check then for my own interest.

Ron Evans

[Heath McKnight]

50%? more like 20-30%. It looked good, because we knew the limitations. Ken, don't sweat it, we get questioned a lot, and I guess it was one of those moments, so no worries! <g>

heath

[Heath McKnight]

I'm re-capturing some clips and will re-do the stills soon.

hwm

[Alan Dunkel]

I was part of a PSA shoot for Film Arts and both a F900 and FX1 were there during the shoot. IMHO, here is what I noticed the most. All the footage that made the final PSA was from the F900.

1. Huge latitude difference and this might be part of the HD "POP". I'm not a pro camera operator and don't discount in the least what Jon and Charles for instance have said here, a pro's experience is a big plus on any shoot with any equipment. Will say the F900 blows the FX1 away as far as latitiude, but the camera cost does too.

2. The Sony FX1 with the expanded LCD focus feature was a charm to focus manually and get a sharp image. The F900 with a 20" Sony studio monitor is probably easy to pull focus on for a pro, but trust me it takes some practice. Good focus is so very important with HD.

Summary, if all we had was the FX1, we could have lighted differently and got a very good result. If you have a choice between the FX1 and the F900 on a shoot and don't mind the extra bulk, much higher rental fee, DUH..... you figure it out.

IMHO, the F900 is in no danger at all from the FX1 or the soon to be released pro version, but the FX1 is a camera I'd like to use more and it is true HD on a budget.

Regards, Alan

[Heath McKnight]

If you have some side-by-side comparison stills, let us know. We'd like to see them!

heath

[Alan Dunkel]

Heath, I do have full resolution F900 footage for the final .30 PSA in Blackmagic 10bit that stills could be pulled from. Unfortunately, there are effects and it has correction applied, so what I have now isn't "raw". The owner of the FX1 is currently in Nepal shooting a documentary with it, so any FX1 stills are unlikely.

Regards, Alan

PS. Would look forward to a head to head review you do between one of the pro format HD cams and the FX1. So what if it can't win. The FX1 is true HD for $3500 with 5 buck tapes and component to my 720p projector had a great look to it.

[Heath McKnight]

I know Jon used my HD10 with a Varicam, so I wouldn't be surprised when Jon shoots with the F900 and asks for the FX1. More soon...

heath

[Charles Papert]

<<<Summary, if all we had was the FX1, we could have lighted differently and got a very good result. >>>

Alan, could you elaborate on what sort of changes you are suggesting for the lighting? Overall level, or contrast (due to latitude issues)?

[Alan Dunkel]

Charles, The lighting was set with the F900 in mind and 24p was used on that camera as well, not for the FX1/1080i. We shot some F900 footage at 1080i toward the end to take a look and that needs more overall light on either camera compared to the 24fps 1080P on the F900. Due to that, I can't really say if the FX1 needs more overall light if both are in 1080i or turning up the gain would do the trick. Personally I'd increase the light level if possible, but the gain on the FX1 seems to be of real use. The FX1 was not as good at handling wide lighting contrast differences and that is what I meant by latitude. Much more of the detail in the shadows could be seen on the F900 footage. Not trying to say it has a problem, we created that contrast with lights, just that the F900 is much superior in that area. The F900 camera, lense, monitor etc. is up in the 100k range and was something like a $1200+ daily rental for the package. Of course I've heard it can be less costly to get a package and DP out of LA for the weekend compared to local rental here. I'm sure film is better yet depending on stock ( ie. not Kodachrome 40, which I've shot and that stock for example probably has less latitude than the FX1 ), but I've only used super8 and 16mm myself. Vision2 on 16mm or S16 for example would handle contrast really well IMHO, but I'm not a tech either. There are certainly times from an artistic standpoint handling of a wider contrast would be a plus or even specifically needed.

For low budget HD work renting a FX1 or ZR1 for a weekend could cost less than a few $100 HDCAM tapes and the editing is much less expensive to do yourself, no HDCAM decks or SDI cards needed ( or $300/hr HDCAM edit suite rental ), just firewire. I was very impressed for what it is.

Regards, Alan