>I finally got the chance to do a side by side comparison of the Panasonic HDX900 and a Sony F900-3 and the results were startling.
>I did the test at PC&E in Atlanta where Panasonic had a representative bring a demo HDX900 and PC&E graciously supplied an F900-3.
>We set both cameras for the normal factory video mode at 1080i for the HDX and 1080p for the F900, white and black balanced and then held a shot of a colour/resolution chart. Right off the bat both cameras were pretty well matched with the F900 being just a bit Sony green. It seems to my eye all Sony cameras and monitors want to be a bit greener than say Panasonic and Ikegami...which tend to be a bit redder... however doing a standard white and black balance took care of that difference.
>Colour and contrast seemed similar to the naked eye as did the blacks.
>In 24p mode, the HDX was a full 2 1/4 stops faster than the F900 with that difference narrowing to 1 full stop at 60i...the Panasonic still being more sensitive.
>In resolution land the F900 had a slight edge. The chart we were using...and I'm sorry but I didn't get the name of the type of chart but it's one most camera houses use, has colour blocks much like a MacBeth but not as many colour blocks...and it has resolution patterns for the horizontal and vertical.
>The vertical resolution pattern starts out with I think 5 or 6 curved lines that converge into a single line. It has numbered stages...1 the widest and somewhere up to 20 as the lines converge up the frame. The Sony held it's resolution to about 8 on the chart with the Panasonic losing that battle at about 7.
>The F900 has 2 million pixels per chip....the Panasonic has 1.4 million pixels per chip so there has to be some difference. Again this was perceived by very close examination of both the Sony HD monitor and the Panasonic monitor we had set up...so we're kind of apples to apples in monitor world.
>I then took both cameras into the parking lot and shot a couple of minutes of video of trees and parking lot stuff...all of which were back lit. Both cameras handled the wide range of contrast quite well and both seemed to capture the same colour, contrast and shaded areas quite well. In fact pictures from both cameras could have been intercut with no perceptual difference.
>I gave the F900 to a friend and we walked back into the building while still rolling. I wanted to see how deep each camera could reach into dark areas. I set both cameras to the "D" filter and with the neutral set to 2 and we walked and rolled back into the building. We passed the through the camera department which is a mix of fluorescent and daylight, then down an unlit hall into camera test area that was only lit by the chart light. That was interesting because we were surely several stops under at that point but both cameras handled the detail in black areas quite well, with the HDX900 doing just a bit better. That's because it's a stop faster in straight video mode.
>My main goal here was to see how well the HDX900 at a price point of $26,500 with no lens, viewfinder, camera mic or base plate, vs the F900-3 which is $80,000, including, viewfinder, camera mic and base plate, sans lens.
>They matched quite well and I believe in a two camera shoot or inter-cutting between the two, it would be difficult to tell the difference.
>I didn't have the chance to do 24p tests between the two but just for the few minutes we went into that mode, the HDX900 was over two stops more sensitive than the F900-3 and the perceptual picture difference was the same, that is to say, they looked very similar.
>The HDX900 has a Varicam chip array that has been placed in an SDX900 body with all the menu and set up functions pretty much the same. The camera does not have variable frame capabilities and does not have the depth in picture manipulation and film emulation the Varicam does but it has enough.
>The HDX900 does everything the SDX900 did only it's HD.
>As a standard feature, the HDX900 comes with a cache card so you can set the pre record from about 2-12 seconds. This also allows for time lapses and allowing for an extremely slow shutter mode.
>Unlike the F900, the HDX900 has a built in downconverter and has firewire out.
>This is going to be a great camera that's going to fill that ENG/EFP niche and it's taped based, which for the immediate future, has a more sensible work flow.
>Within the next few months Panasonic will be coming out with a P2 slotted version of the HDX900. That'll be interesting as we approach the age of tapeless and no moving parts.
>But with that being said, I'm still going to get a couple of the HDX900's because I still think we're 2-4 years away from viable, cost effective tapeless production.
>What monitors were you viewing? LCD's? Unless it is on an ECinema or Cinetal or a BVMD CRT it's really more a monitor demo.
Chief Technology Officer
Band Pro Film & Digital
>Michael Bravin asks about the F900/HDX Demo:" Highlight handling? Playback off tape? Properly painting both cameras? What monitors were you viewing?"
>Yes indeed this was a "demo" comparison. No scopes, engineers or playback decks were involved. No painting, just straight factory sets. Very simple but that's my world.
>I thought each camera handled highlights well, particularly with the outside heavy back lit situations. There were some shots with windows and both cameras with the DCC on, handled that pretty good as well.
>We did not use $10,000 monitors but monitors I would use in the field which is why I had two, Sony and Panasonic monitors and we A/B'd into each. The Panasonic monitor has a grab frame feature which allowed us to view each camera on a split screen on the same monitor.
>My primary interest was in seeing how closely the two camera matched.
>Now if we were blowing to film, I think the differences would be more apparent. I'm not EVEN saying the HDX900 out performed the F900; but for the money and the applications that I WILL be using them for, I was very pleased with the results.
>The camera comparison is also important because I'm doing a resort project in September...one that I shot last year using an F900 but would like to use the HDX900 (because I'll own it) and I will be inter-cutting what I did last year with new material so I wanted to see if that was even possible, and it is...in my opinion.
>Make no mistake. I shoot television, not movies. I live and work in Atlanta where we do HD for TV, commercials and the like. We're not Hollywood but are probably doing what the majority of HD users do so I didn't do as deep a comparison test as those of you involved in doing film outs would probably do.
>The HDX900 is not a film out camera but with that being said, I think it would look pretty good blown to film.
>One more thing about the Demo I did, we used a DSC Labs chart, The Chromo Dumonde, which is standard fare for most camera houses and for monitors we use the Panasonic BTLH 1700W and a Sony LMD 232 WS and A/B'd between them to take into account monitor differential. Also the Panasonic monitor has a frame grab, slit screen feature which allows for side by side comparison of a single image.
class="style15">>>The Panasonic monitor has a grab frame feature which allowed us to >>view each camera on a split screen on the same monitor.
>Any chance we could get a peek at those grabs Allen?
class="style15">>> for monitors we use the Panasonic BTLH 1700W and a Sony LMD >>232 WS and A/B'd between them to take into account monitor >>differential.
>I always enjoy user comments on equipment. Only gotcha to your test was the monitors. I would never use LCD panels for critical camera testing or comparisons unless I had access to an E-Cinema display. A CRT is still the way to go and at least a 20 inch monitor minimum. I prefer Sony BVM's or PVM's in a pinch as I'm not familiar with Panasonic's crt line. However I'm sure Panasonic's top of the line CRT's are excellent also.
>I compared a Varicam to a 900R recently. Other than the 720p resolution hit vs. 1080i the Varicam really impressed me with it's colorimetry and flexibility next to the 900R.
>Joe McDonnell III
Cinematography/ High Definition
New Orleans, La
Los Angeles, Ca
class="style15">>>I would never use LCD panels for critical camera testing or >>comparisons unless I had access to an E-Cinema display.
>I have a Panasonic 1700, and think it's a pretty good field monitor for smaller productions, especially run and guns. I have the Porta-Brace case for it, too. It has a built-in WFM and runs AC/DC (though I still need to find a VESA mount for the Bauer clip). The off-axis viewing is pretty bad, but other than that, it may not be perfect, but it sure is useful. But we're not talking about situations with a proper tent, anyway, so critical viewing isn't possible. But what do you want to know in the field? That a shot's framed properly, in focus, and safe for post, right? It'll tell you those things.
SFD vfx & creative post
Santa Monica, CA
>If both cameras were the same price, which one would you purchase?
>Rick Thompson, hddp
>Rick Thompson asks :
class="style15">>>”If both cameras were the same price, which one would you >>purchase?”
>That is a damn good question!
>I think if the F900R were the same price as the HDX900, one would have to consider that option...but, having shot lots of stuff with an F900-3, I find the camera really heavy, a power suck and noisy particularly when that boy heats up and the fans start kicking in.
>No cache card and no built in downconverter is also stressful but all that aside, if you’re a freelance cameraperson looking for a camera to cover all bases, I still think you have to give the HDX900 serious consideration.
>Now that being said, if the F900R was switchable from 720p to 1080i, that could sway my decision but without the mentioned options and features of the HDX900, I would still be inclined to go the Panasonic route.
>Remember there is the work flow issue as well ; and while HDCAM has a proven, stable edit work flow, the cost of playback decks (of which I need 3) as logging stations is also a deal breaker. While the AJ1400 Panasonic deck costs about the same as the J-3 HDCAM playback deck, the 1400 is a full edit deck with insert editing and full record/play back features. The J-3 is playback only so that fact needs to enter the equation. Deliverables will most certainly always be either DigiBeta or HDCAM or as in my case both and generating an HDCAM master from DCPROHD is a no brainer.
>I think we tend to get emotional about our HD choices. The main thing is to evaluate your entire work flow pattern as well as acquisition needs and factor all that info into making a sound choice that makes production and economic sense.
>This is of course MY opinion but one that I have accepted and am going forward with!
class="style15">>> if you¹re a freelance cameraperson looking for a camera to cover all >>bases, I still think you have to give the HDX900 serious consideration.
>If you are a freelance television cameraman right now the HDX doesn’t cover any of your bases. Its too soon for producers to be asking for it. However they soon will be asking and then the camera will only cover a portion of your HD bases.
>If the F900 did shoot 720P I’m certain very few would use it.
>I prep the F900 and Varicam most days. I see the higher noise levels in the Varicam immediately when comparing it to the F900. I haven’t seen the HDX yet, but if as you say the noise level is comparable, then the HDX is a marked improvement over the Varicam.
>Freelancers rarely choose cameras, producers do. Wish we did!
class="style15">>>I think if the F900R were the same price as the HDX900, one would >>have to consider that option...but, having shot lots of stuff with an F900->>3, I find the camera really heavy, a power suck and noisy …
>The F900R is more like an SX camera or Digibeta in weight. I was impressed how light it is. In an ENG/EFP config it's no heavier than a PDW-530, SX or Digibeta. Weight and balance wise on my shoulder it felt like my old SX ENG camera that I shot with for over 6 years.
class="style15">>>Now that being said, if the F900R was switchable from 720p to 1080.
>Here's an interesting bit of sales gobblygook from Panasonic :
*Multi-format recording system that supports 1080 at 59.94i, 50i, 29.97p, 25p, 23.98p, 23.98pA and 720 at 59.94p, 50p, 29.97p, 25p, 23.98p
>Supports 1080i? Huh? In everyday speak: The HDX900 is a 720p camera performing live active scaling to achieve 1080i or 1080p. Nothing wrong with that except the F900R will blow it out of the water, I guarantee. About the only 3CCD competition the F900R has presently would be the Viper.
>Joe McDonnell III
Cinematography/ High Definition
New Orleans, La
Los Angeles, Ca
>John Chater replies :
class="style15">>>"If you are a freelance television cameraman right now the HDX >>doesn’t cover any of your bases. Its too soon for producers to be >>asking for it."
>I guess I should qualify my statement...for THIS freelancer, it's a no brainer. You are right producers aren't yet calling for it but they will.
>I am a bit unique in that I have a closed loop for broadcast production so I can make the gear work right away and any ancillary work that I used to have rent gear for I'll have my own packages.
>You have to know you market I guess.
>John Chater continues :
class="style15">>>"I see the higher noise levels in the Varicam immediately when >>comparing it to the F900."
>Can't say I disagree but then I see a difference between my 3 BVW600's and my BVW300a but all four cameras are working to the satisfaction of my clients.
>As I said in my original post when comparing the HDX900 with an F900-3, it was perceptual...and to my eye and those who were at the test, the video noise level was more or less equal to the F900/3 but I supposed if you did the test with engineering test gear and a super CRT...yeah, there may be a difference. Non of us noticed it and as I asserted in my previous post, with a price differential of around $50,000 if there's a slight increase in video noise, I'll call it texture and won't even charge extra for it!
>Joe McDonnell adds : "The HDX900 is a 720p camera performing live active scaling to achieve 1080i or 1080p...the F900R will blow it out of the water..."
>No doubt and I never said the HDX900 would be better than either the F900/3 or R series. I mean we're talking 1.4 million pixels vs. 2 million for starters.
>However as I said previously, it's damn good and loaded with features the F900 series just doesn't have and again that price point thingy. I need two HD cameras and there's no way I could afford two HDCAMS. For my needs and perhaps the needs of others, I still assert the HDX900 will perform admirably and those who are used to working with a Varicam should be please
class="style15">>>I see a difference between my 3 BVW600's and my BVW300a but all >>four cameras are working to the satisfaction of my clients.
>That's the bottom line, keeping your clients happy.
>Don't hate me because I'm practical!!!
>I didn't mean to come off that way, sorry. It's the weekend so I'm lighting up a cigar, cracking open an iced Pacifica and kicking back. If only I had some boiled crawfish, crabs and a hot tub. We can resume arguing on Monday...
>Joe McDonnell III
Cinematography/ High Definition
New Orleans, La
Los Angeles, Ca
class="style15">>>Supports 1080i? Huh? In everyday speak: The HDX900 is a 720p >>camera performing live active scaling to achieve 1080i or 1080p. >>Nothing wrong with that except the F900R will blow it out of the water, I >>guarantee.
>You are correct about the HDX900 scaling up to 1080, but I don't think the F900R will "blow it out of the water". I will know more next week when I have a chance to beat extensively on an HDX900 under controlled conditions, using large real CRT monitors and scopes. I have no intention of comparing the tape formats - this is strictly a camera evaluation, and strictly for my own benefit, because I'm in the market.
>I want to see how well I can pull mattes with the Panasonic.
>It is likely the F900R will be better in some respects, but for three times the price of the HDX900, I would expect to see HUGE differences between them, and that's not likely to happen. In fact, having informally seen the images from the HDX900, my guess is that five feet back from a 24" Sony CRT monitor, there are fewer than a dozen people on this list who would be able to consistently pick out which is which.
class="style15">>> It is likely the F900R will be better in some respects, but for three >>times the price of the HDX900, I would expect to see HUGE >>differences between them, and that's not likely to happen.
>I used to say that the difference between cameras was roughly equal to the ratio of the square root of the prices. These days it may be closer to the cube root...
class="style15">>>It is likely the F900R will be better in some respects, but for three >>times the price of the HDX900, I would expect to see HUGE >>differences between them, and that's not likely to happen.
>Here I am replying to my own post...
>So I spent about four hours yesterday comparing an F900/3 and an HDX900, shooting a small tabletop greenscreen setup with Barbie and Midge (in some of the latest fall fashions) as models. The comparison was conducted at Wexler Video in Burbank, who generously supplied the location, lights, an F900/3, a 24" Sony HD monitor, a Leader 5750 scope, the lenses, and various miscellaneous bits and pieces. The HDX900 was supplied by Doug Leighton of Panasonic. It was a pre-shipment sample, not a production unit, but Doug assured me it was very close to what would ship (or is shipping now, I understand).
>Since colour screen work is pretty much what I do, and under generally controlled conditions, I spent most of my time looking at the cameras from that perspective. It's not that important to me how they handle highlights out a blown out window, for example. I'm much more interested in resolution, colorimetry, and sensitivity
>The cameras had identical lenses (Canon 22x), and I set each one up to taste, using DSC Labs' excellent ChromaDuMonde chart, the Sony 24" HD monitor, and Leader 5750 Waveform/Vectorscope. How anybody can set up a camera (especially the matrices) properly without at least that chart and a waveform/vectorscope is beyond me. Having the 24" monitor was a big luxury. I started off with both cameras in 1080i 59.94 mode.
>The HDX900 required quite a bit of adjustment (I didn't like the way it looked out of the box at all), but the good news is that all the handles I needed to make it look how I wanted were there. The menu structure is similar to the Varicam, although not as extensive, and it only took me a few minutes to get comfortable with what was available.
>So once I had them looking the way I wanted, I put both through an HD Ultimatte and looked at the results. Again, this is *strictly* an evaluation of the camera parts, not the recorders.
>The first thing I noticed was how much quieter the HDX900 was. It was a LOT quieter, and that was clearly evident in the shadow noise through the Ultimatte - the HDX900 didn't have *any*, while the F900/3 had quite a bit. It was also evident in the lack of noise in the black wardrobe (I had switched the girls to a basic black felt burka evening look to test for noise and edging). There was some minimal edging on the HDX900 (more on the F900/3), but nothing the Ultimatte couldn't deal with.
>The HDX900 tracked grey scale well, the matrix setups were useful, the built in gamma curves were OK, although I didn't spend a lot of time looking at them because I don't care. Dynamic range seemed to be similar to the F900/3, with neither being all that fabulous if you don't torture the image with knees.
>I was quite pleased with the detail adjustments available in the HDX900 - there were enough handles to make the image look good without getting edgy or excessively noisy. Unless you're running with detail off, I have found the detail setup on most cameras to be the single most screwed up part. You're looking to enhance apparent resolution, not sharpness or edges. I know some here leave detail off as a matter of course, but if you're going to be downconverting and using that downconverted output for money, that's just not how I would go. You want to add detail when you're still working with original full rez RGB, not later in the downconversion process or post.
>The cameras were roughly the same speed, although some of the built-in gamma curves made the HDX900 about half a stop slower.
>The cross conversion from 720 to 1080 on the HDX900 looked surprisingly good. I would have no problem using it for colour screen work in either mode. The camera needs to be rebooted when switching modes, of course.
>Unlike the F900/3, the HDX900 always inserts 3:2 into the HDSDI output, so while the F900 puts out 48i when in 24p, the HDX900 puts out 59.94i no matter what mode it's in, so without going through some sort of post process, there was no real point for me in comparing them at 24p, since they would be so visibly different.
>I really wanted to see the remote paintbox, but Doug said it wasn't yet available. About $4.5K when it is (he said), which I consider to be out of proportion, especially since the HDX900 is not compatible with previous versions of the remote paintbox. Half to two-thirds that price would have been a lot more reasonable.
>Overall, I was very impressed with the HDX900, especially at the price point. I do have to say that Panasonic touting the list price of the camera without a viewfinder or mounting plate is a bit, umm, chickenshit. Oh, did you actually want to be able to put the camera on a tripod, frame the shot, and focus? That's extra. Since no one in the known universe is going to buy the camera without a viewfinder or mounting plate, think around $30K list price, not $26.5K.
>"Did you want wheels with that Toyota?" aside, I've moved the HDX900 to the top of my current list when I'm ready to commit to an HD camera. Nice job, Panasonic. And I stand by a slightly modified version of my earlier statement (since I used an F900/3 and not the newer F900R) that once set up properly, five feet back from a 24" Sony CRT monitor, very very few people would be able to consistently pick out which is which.
>Disclaimer : I am not affiliated with Panasonic or any of the manufacturers mentioned in any way except as a paying customer blah blah blah. Wexler does employ me from time to time when no one else is available. Barbie and Midge were not harmed in any way during these tests.