Tag Archives: test

Is This The Age Of The Small Camera Part 2

This is part 2 of my 2 part look at whether small cameras such as a Sony FX3 or A1 really can replace full size cinema cameras.

For this part of the article to make sense you will want to watch the YouTube clips that are linked here full screen at at the highest possible quality settings, Preferably 4K. Please don”t cheat, watch them in the order they are presented as I hope this will allow you to understand the points I am trying to make better.

Also, in the videos I have not put the different cameras that were tested side by side. You may ask why – well it’s because if you do watch a video online or a movie in a cinema you don’t see different cameras side by side on the same screen at the same time. A big point of all of this is that we are now at a place where the quality of even the smallest and cheapest  large sensor camera is likely going to be good enough to make a movie. It’s not necessarily a case of is camera A better than camera B, but the question is will the audience know or care which camera you used. There are 5 cameras and I have labelled them A through to E.

The footage presented here was captured during a workshop I did for Sony at Garage Studios in Dubai (if you need a studio space in Dubai they have some great low budget options). We weren’t doing carefully orchestrated  camera tests, but I did get the chance to quickly capture some side by side content.

So lets get into it.

THE FINAL GRADE:

In many regards I think this is the most important clip as this is how the audience would see the 5 cameras. It represents how they might look at the end of a production. I graded the cameras using ACES in DaVinci Resolve. 

Why ACES? Well, the whole point of ACES is to neutralise any specific camera “look”.  The ACES input transform takes the cameras footage and converts it to a neutral look that is meant to represent the scene as it actually was but with a film like highlight roll off added. From here the idea is that you can apply the same grade to almost any camera and the end result should look more or less the same. The look of different cameras is largely a result of differences in the electronic processing of the image in post production rather than large differences in the sensors. Most modern sensors capture a broadly similar range of colours with broadly similar dynamic range. So, provided you know the what recording levels represent what colour in the scene, it is pretty easy to make any camera look like any other, which is what ACES does.

The footage captured here was captured during a workshop, we weren’t specifically testing the different cameras in great depth. For the workshop the aim was to simply show how any of these cameras could work together. For simplicity and speed I manually set each camera to 5600K and as a result of the inevitable variations you get between different cameras, how each is calibrated and how each applies the white balance settings there were differences between in the  colour balance of each camera.

To neutralise these white balance differences the grading process started by using the colour chart to equalise the images from each camera using the “match” function in DaVinci Resolve. Then each camera has exactly the same grade applied – there are no grading differences, they are all graded in the same way.

Below are frame grabs from each camera with a slightly different grade to the video clips, again, they all look more or less the same.

A-graded_1.1.11-600x316 Is This The Age Of The Small Camera Part 2
The graded image from camera A. Click on the image to view the full resolution image.

 

B-graded_1.2.1-600x316 Is This The Age Of The Small Camera Part 2
The graded image from camera B. Click on the image to view the full resolution image.

 

C-graded_1.3.4-600x316 Is This The Age Of The Small Camera Part 2
The graded image from camera C. Click on the image to view the full resolution image.

 

D-Graded_1.4.1-600x316 Is This The Age Of The Small Camera Part 2
The graded image from camera D. Click on the image to view the full resolution image.

 

E-graded_1.5.1-600x316 Is This The Age Of The Small Camera Part 2
The graded image from camera E. Click on the image to view the full resolution image.



The first thing to take away from all of this then is that you can make any camera look like pretty much any other and a chart such as the “color checker video” and software that can read the chart and correct the colours according to the chart makes it much easier to do this.

To allow for issues with the quality of YouTube’s encoding etc here is a 400% crop of the same clips:

 

What I am expecting is that most people won’t actually see a great deal of difference between any of the cameras. The cheapest camera is $6K and the most expensive $75K, yet it’s hard to tell which is which or see much difference between them. Things that do perhaps stand out initially in the zoomed in image are the softness/resolution differences between the 4K and 8K cameras, but in the first un cropped clip this difference is much harder to spot and I don’t think an audience would notice especially if the one camera is used on it’s own so the viewer has nothing to directly compare it with. It is possible that there are also small focus differences between each camera, I did try to ensure each was equally well focussed but small errors may have crept in.

WHAT HAPPENS IF WE LIFT THE SHADOWS?

OK, so lets pixel peep a bit more and artificially raise the shadows so that we can see what’s going on in the darker parts of the image.

 

There are differences, but again there isn’t a big difference between any of the cameras. You certainly couldn’t call them huge and in all likelihood, even if for some reason you needed to raise or lift the shadows by an unusually large amount as done here (about 2.5 stops) the difference between “best” and “worst” isn’t large enough for it to be a situation where any one of these cameras would be deemed unusable compared to the others.

SO WHY DO YOU WANT A BETTER CAMERA?

So, if we are struggling to tell the difference between a $6K camera and a $75K one why do you want a “better” camera? What are the differences and why might they matter?

When I graded the footage from these cameras in the workshop it was actually quite difficult to find a way to “break” the footage from any of them. For the majority of grading processes that I tried  they all held up really well and I’d be happy to work with any of them, even the cameras using the highly compressed internal recordings held up well. But there are differences, they are not all the same and some are easier to work with than the others. 

The two cheapest cameras were a Sony FX3 and a Sony A1. I recorded using their built in codecs, XAVC-SI in the FX3 and XAVC-HS in the A1. These are highly compressed 10 bit codecs. The other cameras were all recorded using their internal raw codecs which are either 16 bit linear or 12 bit log. At some time I really do need to do a proper comparison of the internal XAVC form the FX3 and the ProResRaw that can be recorded externally. But it is hard to do a fully meaningful test as to get the ProResRaw into Resolve requires transcoding and a lot of other awkward steps. From my own experience the difference in what you can do with XAVC v ProResRaw is very small.

One thing that happens with most highly compressed codecs such as H264 (XAVC-SI) or H265(XAVC-HS) is a loss of some very fine textural information and the image breaking up into blocks of data. But as I am showing these clips via YouTube in a compressed state I needed to find a way to illustrate the subtle differences that I see when looking at the original material. So, to show the difference between the different sensors and codecs within these camera I decided to pick a colour using the Resolve colour picker and then turn that colour into a completely different one, in this case pink.

What this allows you to see is how precisely the picked colour is recorded and it also shows up some of the macro block artefacts. Additionally it gives an indication on how fine the noise is and the textural qualities of the recording. In this case  the finer the pink “noise” the better, as this is an indication of smaller, finer textural differences in the image. These smaller textural details would be helpful if chroma keying or perhaps for some types of VFX work. It might (and say might because I’m not convinced it always will) allow you to push a very extreme grade a little bit further.

I would guess that by now you are starting to figure out which camera is which – The cameras are an FX3, A1, Burano, Venice 2 and an ArriLF.

In this test you should be able to identify the highly compressed cameras from the raw cameras. The pink areas from the raw cameras are finer and less blocky, this is a good representation of the benefit of less compression and a deeper bit depth.

A-Compression_1.1.1-600x316 Is This The Age Of The Small Camera Part 2
Camera A. Click on the image to view the full resolution image.

 

B-Compression_1.2.1-600x316 Is This The Age Of The Small Camera Part 2
Compression and codec Camera B. Click on the image to view the full resolution image.

 

C-Compression_1.3.1-600x316 Is This The Age Of The Small Camera Part 2
Compression and codec Camera C. Click on the image to view the full resolution image.

 

D-compression_1.4.1-600x316 Is This The Age Of The Small Camera Part 2
Compression and codec Camera D. Click on the image to view the full resolution image.

 

E-Compression_1.4.1-600x316 Is This The Age Of The Small Camera Part 2
Compression and codec Camera E. Click on the image to view the full resolution image.



But even here the difference isn’t vast. It certainly, absolutely, exists. But at the same time  you could push ANY of these cameras around in post production and if you’ve shot well none of them are going to fall apart. 

As a side note I will say that I find grading linear raw footage such as the 16 bit X-OCN from a Venice or Burano more intuitive compared to working with compressed Log. As a result I find it a bit easier to get to where I want to be with the X-OCN than the XAVC. But this doesn’t mean I can’t get to the same place with either.

RESOLUTION MATTERS.

Not only is compression important but so too is resolution. To some degree increasing the resolution can make up for a lesser bit depth.  As these camera all use bayer sensors the chroma resolution will be somewhat less than the luma resolution. A 4K sensor such as the one in the FX3 or the Arri LF will have much lower chroma resolution than the 8K A1, Burano or Venice 2. If we look at the raised shadows clip again we can see some interesting things going on the the girls hair.

 

If you look closely camera D has a bit of blocky chroma noise in the shadows. I suspect this might be because this is one of the 4K sensor cameras and the lower chroma resolution means the chroma noise is a bit larger.

I expect that by now you have an idea of which camera is which, but here is the big reveal: A is the FX3, B is the Venice 2, C is Burano, D is an Arri LF, and E is the Sony A1.

What can we conclude from all of this: 

There are differences between codecs. A better codec with a greater bit depth will give you  more textural information. It is not necessarily simply that raw will always be better than YUV/YCbCr but because of raws compression efficiency it is possible to have very low levels of compression and a deep bit depth. So, if you are able to record with a better codec or greater bit depth why not do so. There are some textural benefits and there will be fewer compression artefacts. BUT this doesn’t mean you can’t get a great result from XAVC or another compressed codec.

If using a bayer sensor than using a sensor with more “K” than the delivery resolution can bring textural benefits.

There are differences in the sensors, but these differences are not really as great as many might expect. In terms of DR they are all actually very close, close enough that in the real world it isn’t going to make a substantial difference. As far as your audience is concerned I doubt they would know or care. Of course we have all seen the tests where you greatly under expose a camera and then bring the footage back to normal, and these can show differences. But that’s not how we shoot things. If you are serious about getting the best image that you can, then you will light to get the contrast and exposure that you want. What isn’t in this test is rolling shutter, but generally I rarely see issues with rolling shutter these days. But if you are worried about RS, then the Venice 2 is excellent and the best of the group tested here.

Assuming you have shot well there is no reason why an audience should find the image quality from the $6K FX3 unacceptable, even on a big screen. And if you were to mix and FX3 with a Venice 2 or Burano, again if you have used each camera equally well I doubt the audience would spot the difference.

BACK TO THE BEGINNING:

So this brings me back to where I started in part 1. I believe this is the age of the small camera – or at least there is no reason why you can’t use a camera like an FX3 or an A1 to shoot a movie. While many of my readers I am sure will focus on the technical details of the image quality of camera A against camera B, in reality these days it’s much more about the ergonomics and feature set as well as lens and lighting choices.

A small camera allows you to be quick and nimble, but a bigger camera may give you a lot more monitoring options as well as other things such as genlock. And….. if you can – having a better codec doesn’t hurt. So there is no – one fits all – camera that will be the right tool for every job.  

Testing the Sony ILME-FX30

 

Please watch the video to see my video review or read on:

A few weeks ago I borrowed an FX3 from Sony for some testing in order to better understand the performance of this budget Cinema Line camera. I used it over a long weekend to shoot some circus acts and to perform some basic tests. By the end of the weekend of testing I decided to get one for myself even though I already own an FX3 and FX6. 

Screenshot-2022-12-06-at-16.33.35-copy-600x335 Testing the Sony ILME-FX30
I shot various circus acts with the FX30.

 

So what made me buy the FX30?

For a start it’s cheap. At around $2000 for the body only you get a lot of camera for the money. If you want the same handle as the FX3 with XLR inputs, add another approx $500 to the base price. But as well as the low price I also I really like the fact that it is super 35 rather than full frame. The FX30’s 6K APSC sized sensor delivers really good oversampled 4K from a scan area very similar to super 35mm film. This means you can use it with almost any classic cinema lens, of which there are many to choose from. You can use it with zoom lenses designed for s35 (again which there are many to choose from) as well as lower cost APSC lenses.  A combination that I am particularly fond of is the FX30 plus the Sony 18-105mm f4 G APSC power zoom. While this combination isn’t ever going to win an award for the ultimate in image quality it is very reasonable.  It gives me great look images at a wide range of focal lengths in a surprisingly small package. 

But just how good is the image quality?

Sony advertise the FX6 and FX3 as having 15+ stops of dynamic range, while only claiming 14+ stops for the FX30. So one of the first tests that I did was to compare the dynamic range of both the FX6 and FX30 using my home made dynamic range tester. While this device isn’t necessarily ultra accurate, it is consistent and it allows me to visually compare the DR of the two cameras. I also thought it would be interesting to include the FS7, another s35 camera in my tests.

Screenshot-2022-12-06-at-15.24.32-copy-600x231 Testing the Sony ILME-FX30
Dynamic range test to compare the FX30 to the FX6

As you can see from the above image, the dynamic range of the FX30 is extremely close to that of the FX6, so close in fact that I was unable to measure a difference with my home made tester. There is a 15th stop buried deep in the noise of both cameras and at 800 ISO the noise is very similar from both camera, if anything, visually I prefer the look of the very fine noise from the  FX30, probably a result of the 6K over sampling.

But what about compared to the FS7? In this image you can see how in the shadows the FS7 produces a lot of coloured chroma noise compared to the FX30. It is this chroma noise that makes it desirable to expose the FS7 a bit brighter than Sony’s base recommendation as it is quite distracting in lower exposures. So against the FS7, for me the FX30 is a clear winner in the dynamic range stakes.

Screenshot-2022-12-06-at-15.25.34-copy-600x302 Testing the Sony ILME-FX30
Coloured noise in the shadows of the FS7 limit the useable shadow range compared to the FX30/FX3/FX6. In a video sequence the FS7’s noise is very obvious. Click on the image to enlarge it.

What about resolution?

OK, so the FX30 does not lack dynamic range, what about resolution, how does it compare with the FX6? To see this image larger please click on it. And be aware that scaling of the image that may be happening in your browser or computer and that scaling may add aliasing and moire to the images not in the original.

Screenshot-2022-12-06-at-16.02.50-600x344 Testing the Sony ILME-FX30
Comparing the resolution of the FX30 and the FX6. Click on the image to enlarge it.

 

What you can see from the above test is that aliasing starts to occur at a slightly lower resolution for the FX6 than the FX30. Aliasing happens when the resolution of the image falling on the sensor exceeds the  resolving power of the sensor.  This result isn’t really a surprise, the FX6 like the FX3 has a sensor that is a little over 4K pixels wide and it would appear that Sony tuned the optical filtering to squeeze as much resolution from this sensor as possible. Meanwhile the FX30 has a 6K pixel wide sensor, so it is easier to get close to 4K resolution without excessive aliasing. 

We can also see a difference in the coloured moire of these two cameras.

Screenshot-2022-12-06-at-15.43.11-copy-600x345 Testing the Sony ILME-FX30
The FX6 produces more moire and aliasing than the FX30, click on the image to enlarge.

 

And I also chose to test the FS7 to see how much moire the FS7 produced. The FS7 was the worst of the 3 cameras by some way with a fair amount of strong coloured moire.

Screenshot-2022-12-06-at-15.48.40-copy-600x345 Testing the Sony ILME-FX30
The FS7 produces more moire and aliasing than both the FX30 and the FS7, click on the image to enlarge.

 

I think what we are seeing here is simply improvements in the design of the Optical Low Pass Filter (OLPF) combined with the oversampled 6K sensor of the FX30 delivering an improvement in both resolution and moire/aliasing performance. The FX30 is a camera that is 8 years younger than the FS7, so you would hope that it would be better.

So, in the resolution stakes, the FX30 wins against the FS7, FX6 and FX3.

What about low light performance?

The FX30 has a Dual Base ISO sensor with 2 base ISO’s when shooting S-Log3 of 800 and 2500 ISO. The performance at these 2 ISO is very similar. The dynamic range is broadly the same and the noise is similar. But I would not say the noise is the same, there is more noise at 2500 than there is at 800, but not significantly more.

On the other hand the FX6 has a dual sensitivity sensor and its two base ISO’s are 800 and 12,800. This is a huge difference. You would need to add 24dB of gain to get from 800 ISO to 12,800 ISO and while the 12,800 base is noticeably noisier than the 800 ISO base, it is still quite useable. There is a small reduction in dynamic range at 12,800, but it isn’t really significant.

If you need to shoot in very, very low light the FX6 and FX3 are the clear winners, they are more sensitive than the FX30. But the FX30 isn’t as far behind as you might think. The 6K to 4K oversampling means the noise grain is very fine, so even with a bit of extra gain added in post production to bring it up to the equivalent of 12,800 ISO it doesn’t look terrible. It’s clearly not as good as the FX6, but if you needed to shoot in very low light the FX30 isn’t going to be a complete disaster.

First the FX6:

Screenshot-2022-12-06-at-16.02.31-600x333 Testing the Sony ILME-FX30
FX6 shot at 12,800.

And then the FX30, shot using the exact same light levels and exposure using 2500 ISO and then graded to match the FX6 which was at 12,800 ISO.

Screenshot-2022-12-06-at-16.02.50-copy-600x333 Testing the Sony ILME-FX30
FX30 shot at 2500 ISO then graded to match the FX6. Same light level and exposure as the FX6.

 

I recommend you watch the video review to see these frames larger. There is more noise in the final FX30 image, but it’s not as far from the FX6 as you might imagine. But, on the sensitivity stakes, the FX6/FX3 are without doubt the winners.

What about colour matching?

A couple of quick tests, done both with S-Cinetone and S-Log3 confirmed what I expected I would find. As the FX30 is a part of Sony’s Cinema Line it looks pretty much like every other Cinema Line camera. The colours are extremely close to the FX6. It’s not totally identical, There are some very, very small differences. You do need to match the white balance of both as if you dial the same preset into both the colour temperature of each will be a little off, but once you find the matching white balance the images each produces will be close enough that only close side by side, like for like examination will reveal the subtle differences that do exist. I certainly have no concerns over using both the Fx30 and FX6 on the same shoot. 

Screenshot-2022-12-06-at-16.15.51-copy-600x214 Testing the Sony ILME-FX30

What else do I need to know?

The FX30 does have more rolling shutter than the FX6, but it really isn’t terrible, it’s little different to the FS7. I suggest you watch the video and look at the circus footage that I shot with the FX30, rolling shutter didn’t cause me any issues.

The one thing that the FX30 does exhibit is a little bit of image smear. This occurs when you have a very bright highlight against a very dark background. What you get is a brightening of the background in line with the bright highlight. The FX6 isn’t totally smear free, but it’s very difficult to see the smear on the FX6, it’s not quite so hard to find it on the FX30. But for the vast majority of real world applications I doubt this will cause any major concerns, it certainly didn’t spoil any of my circus footage which often included very bright lights agains dark backgrounds.

Screenshot-2022-12-06-at-16.22.40-copy-600x197 Testing the Sony ILME-FX30
FX30 CMOS smear (circled in yellow)

As you can see, even when looking for it, it isn’t always obvious.

In Conclusion.

Both practically and technically I really like the FX30. Mine will be used on my gimbal with the 18-105 zoom or handheld as a pocket sized camera (yeah, OK, a very big pocket). It has all the same codecs as the FX3 and it has breathing compensation, a fine step variable shutter (similar to ECS shutter) and you can use it as a very high quality webcam. It has the same CineEI modes as the FX3 plus an additional CineEI mode that allows you to add gain to the S-Log3 recordings.

Technically it performs really well. It has great DR and delivers a high resolution image with very well controlled aliasing and moire. Skin tones look great, full of subtle and fine textures. It’s plenty sensitive enough for most normal applications thanks to it’s two base ISO’s of 800 and 2500 (for S-Log3) and the colours extremely closely match those of the FX6, FX3 and FX9.

For the money, the FX30 is a lot of camera. 

CineD Venice 2 Dynamic Range Tests

The recent publication of CineD’s Venice 2 lab tests has created quite a stir and many have asked what my view on this is. You can see the entire test here:  https://www.cined.com/sony-venice-2-lab-test-rolling-shutter-dynamic-range-and-latitude/

I have not done any formal dynamic range testing with Venice 2 myself, but I have shot with it several times. I have also shot with most of Sony’s recent cameras including the original Venice, many different Red cameras and Arri ALexa’s. 

Whenever I shot with Venice 2 the dynamic range has always impressed me. I have been able to pull lots of detail out of the deepest shadows without any issue, no nasty noise artefacts, no coloured blotches.  When I shot the “London Vistas” video in London at night using available light I found the cameras noise floor to be very low, allowing me to get deep shadow textures without issue. The cameras highlight handling has also always impressed me and every time I’ve used Venice 2 I have been delighted with the dynamic range it delivers, it is up there with the Arri Alexa. From my real world shooting experience Venice 2 delivers more DR than my FX9 or FX6 and it delivers it in a very pleasing way. The way the far highlights and deep shadows behave is beautiful.

I would also point out that there are many great examples of deep shadow details and textures that are colour blotch free in Rob Hardy’s “Venizia” short film.

I would point out that CineD noted that the Venice 2 as the delivered the second highest dynamic range result they have seen in their lab when they recorded using the internal 4K ProRes recordings.  Venice 2 comes in just 0.3 stops behind the Alexa in this mode in the CineD tests. CineD have put this down to downsampling from 8K plus the use of additional internal noise reduction. While DCT codecs like ProRes do normally incorporate some degree of NR, I doubt that Sony are doing any significant NR in camera as this tends to degrade the image in other areas. So I find the discrepancy between the results they are seeing between the 16 it X-OCN and the 10 bit ProResHQ very intriguing and it makes me wonder if something else is going on. Downsampling from 8K will certainly help lower the noise a little, but I feel that there is something odd with the X-OCN results, one thing I note is a very raised pedestal on the waveform of the X-OCN, which is somewhat odd, the bit depth should help separate the noise from the useable signal. A camera either has a dynamic range or it doesn’t, only rarely does NR make a significant difference as the sensor analog to digital converters tend to be the one of the main limiting factors. My own real world experience is that Venice 2 when shooting X-OCN has more useable DR than almost every other camera I have used.

Bottom line is – don’t go by the test, try the camera for yourself as I am quite sure you will find, like me, that one thing Venice 2 does not lack is dynamic range. I will try to do my own formal tests as soon as possible.

Camera Test Charts To Print At Home

Screenshot-2022-05-16-at-11.11.38-1024x622 Camera Test Charts To Print At HomeWhen testing and evaluating a camera, whether that’s a digital photo camera, video camera or digital cinema camera it is always useful to have a test chart or 2 (or more). While printing a chart at home isn’t always the best way to go, comercial charts can be very expensive to buy. So below is a link to an ISO 12233 chart, a Zone Plate chart and a Siemens Star chart that you can download  for free and print at home. You will need a good printer and good quality photo paper for the best results. 

For the ISO 12233 chart I divide this into quarters, print each 1/4 and then join them back together to make a larger chart.

The zone plate and siemens charts should be printed as large as possible, but in use they would not fill the frame, perhaps only a small part of the frame depending on the resolution of the camera you are evaluating.

I have not included any color charts or grey scale charts as it will be extremely difficult to know whether the colors or shades of grey your printer produces are actually correct, making the chart invalid.

Click on the green link or the images below:

Click here for ISO 12233 Test Chart as at the top of this page.

Zone-Small-150x150 Camera Test Charts To Print At Home
Zone plate derived from Imatest Software. Click on the image to go to full size file.
Siemens-100-chart-150x150 Camera Test Charts To Print At Home
Siemens 100 radial star chart – click on the image to go to the full size file.

FX9 ISO Rating Confirmation Test

While I had the light meter and exposure test chart out for the FX6 I decided to do the same exposure level confirmation test for the FX9. No nasty surprises, the FX9’s ISO ratings certainly appear to be correct. Again using a DSC Labs exposure reference chart with 18% middle grey and 90% white plus my trusty Sekonic I tested the FX9 at both 800 ISO and 4000 ISO and my light meter and the camera were in good agreement. At 800 ISO the light meter was saying f4.01 while the camera was at f4, I suspect this tiny difference is probably down to transmission losses in the lens.

FX9-Exposure-Test-800_1.2.1 FX9 ISO Rating Confirmation Test
FX9 Exposure rating test at 800 ISO.
FX9-Exposure-Test-4000_1.1.1 FX9 ISO Rating Confirmation Test
FX9 Exposure ISO rating test at 4000 ISO.

FX6 ISO rating Confirmation Test

I have already done this a few times, but having seen some other tests suggesting the FX6’s ISO ratings were incorrect. So I decided to re-confirm my previous findings, which is that the ratings Sony give their cameras is correct. For the test I used a DSC labs exposure calibration chart which is an extremely accurate 18%/90% reflectivity chart and my trusty Sekonic light meter. As you can see at both 800 ISO and 12,800 ISO the light meters indicated exposure settings perfectly match the camera’s ISO ratings, shutter speed and aperture. For the 12,800 ISO test, as my light meter doesn’t go up to 12,800 ISO  I set the light meter to 6400 ISO which is one stop lower than the cameras 12,800. The light meter indicated f11 which is one stop below the f16 required by the camera – confirming that the ISO rating is correct.

FX6-Exposure-test-800_1.2.1 FX6 ISO rating Confirmation Test
800 ISO FX6 exposure calibration test.
FX6-Exposure-test-12800_1.1.2 FX6 ISO rating Confirmation Test
FX6 12,800 ISo exposure calibration test.

ACS Technical Panel Review The PXW-FX9

The ACS have produced a video report about some of the testing that they did with a pre-production FX9. It’s quite a long video but has some interesting side by side comparisons with the FS7 which we all already know very well. You’ve heard much of what’s in the video from me already, but I’m a Sony guy, so it’s good to hear the same things from the much more impartial ACS.

With my super geek hat on it was really interesting to see the colour response tests performed by Pawel Achtel ACS at 37.08. These tests use a very pure white light source that is split into the full spectrum and then the monochromatic light is projected onto the sensor. It’s a very telling test. I was quite surprised to see how large the FS7’s response is, it’s not something I have ever had the tools to measure. The test also highlights a lack of far red response from the FS7. It’s not terrible, but does help explain why warm skin tones perhaps don’t always look as nice as they could. I do wonder if this is down to the characteristics of the cameras IR cut filter as we also know the sensor to be quite sensitive to IR. The good news is that the PXW-FX9 has what Pawel claims to be the best colour accuracy of any camera he’s tested, and he’s tested pretty much all of the current cinema cameras. Take a look for yourself.

Sony FX9 ACS Roundtable from ACS on Vimeo.

Guess The Lens! A little bit of fun and an interesting test.

Last week I was at O-Video in Bucharest preparing for a workshop the following day. They are a full service dealer. We had an FX9 for the workshop and they had some very nice lenses. So with their help I decided to do a very quick comparison of the lenses we had. I was actually very surprised by the results. At the end of the day I definitely had a favourite lens. But I’m not going to tell you which one yet.

The 5 lenses we tested were: Rokinon Xeen, Cooke Panchro 50mm, Leitz (lecia) Thalia, Zeiss Supreme Radiance and the Sony 28-135mm zoom that can be purchased as part of a kit with the FX9.

I included a strong backlight in the shot to see how the different lenses dealt with flare from off-axis lights. 2 of the lenses produced very pronounced flare, so for those lenses you will see two frame grabs. One with the flare and one with the back light flagged off.

I used S-Cinetone on the FX9 and set the aperture to f2.8 for all of the lenses except the Sony 28-135mm. For that lens I added 6dB of gain to normalise the exposure, you should be able to figure out which of the examples is the Sony zoom.

One of the lenses was an odd focal length compared to all the others. Some of you might be able to work out which one that is, but again I’m not going to tell you just yet.

Anyway, enjoy playing guess the lens. This isn’t intended to be an in depth test. But it’s interesting to compare lenses when you have access to them.  I’ll reveal which lens is which in a couple of weeks in the comments. You can click on each image to enlarge it.

Big thanks to everyone at O-Video Bucharest for making this happen.

Lens1-flare Guess The Lens! A little bit of fun and an interesting test.
Lens 1 with flare from backlight.
lens1-no-flare Guess The Lens! A little bit of fun and an interesting test.
Lens 1 with backlight flagged to reduce the flare.
lens2 Guess The Lens! A little bit of fun and an interesting test.
Lens 2
lens-3 Guess The Lens! A little bit of fun and an interesting test.
Lens 3
lens-4-no-flare Guess The Lens! A little bit of fun and an interesting test.
Lens 4
lens-5-flare Guess The Lens! A little bit of fun and an interesting test.
Lens 5 with flare from backlight
lens-5-No-flare Guess The Lens! A little bit of fun and an interesting test.
Lens 5 with backlight masked to kill the flare.

The PXW-FX9 in the real world.

There are already a few setup and staged video samples from the new Sony PXW-FX9 circulating on the web. These are great. But how will it perform and what will the pictures look like for an unscripted, unprepared shoot? How well will the autofocus work out in the street, by day and by night? How does the S-Cinetone gamma and colour in custom mode compare with S-Log3 and the s709 Venice LUT compare?

To answer these questions I took a pre-production FX9 into the nearby town of Windsor with a couple of cheap Sony E-Mount lenses. The lenses were the Sony 50mm f1.8 which costs around $350 USD and the 28-70mm f3.5-f5.6 zoom that costs about $400 USD and is often bundled as a kit lens with some of the A7 series cameras.

To find out how good the auto focus really is I decided to shoot entirely using auto focus with the AF set to face priority. The only shot in the video where AF was not used is the 120fps slow-mo shot of the swans at 0:53 as AF does not work at 120fps.

Within the video there are examples of both S-Cinetone and S-Log3 plus the s709 LUT. So you know which is which I have indicated this is the video. I needed to do this as the two cut together really well. There is no grading as such. The S-Cinetone content is exactly as it came from the camera. The CineEI S-Log3 material was shot at the indicated base ISO and EI, there was no exposure offset. In post production all I did was add the s709 LUT, that’s it, no other corrections.

The video was shot using the Full Frame 6K scan, recording to UHD XAVC-I.

For exposure I used the cameras built in waveform display. When in CineEI I also used the Viewfinder Gamma Display assist function. Viewfinder Gamma assist gives the viewfinder the same look as the 709(800) LUT. What’s great about this is that it works in all modes and at all frame rates. So even when I switched to 2K Full Frame scan and 120fps the look of the image in the viewfinder remained the same and this allowed me to get a great exposure match for the slow motion footage to the normal speed footage. 

AUTOFOCUS.

There are some great examples of the way the autofocus works throughout the video. In particular the shot at 0:18 where the face priority mode follows the first two girls that are walking towards the camera, then as they exit the frame switches to the two ladies following behind without any hunting. I could not have done that any better myself. Another great example is at 1:11 where the focus tracks the couple walking towards the camera and once they exit the shot the focus smoothly transitions to the background. One of the nice things about the AF system is you can adjust the speed at which the camera re-focusses and in this case I had slowed it down a bit to give it a more “human” feel.

Even in low light the AF works superbly well. At 1:33 I started on the glass of the ornate arch above the railway station and panned down as two people are walking towards me. The camera took this completely in it’s stride doing a lovely job of shifting the focus from the arch to the two men. Again, I really don’t think I could have done this any better myself.

NOISE.

Also, I am still really impressed by how little noise there is from this camera. Even in the high ISO mode the camera remains clean and the images look great. The low noise levels help the camera to resolve colour and details right down into the deepest shadows. Observe how at 2:06 you can clearly see the different hues of the red roses against the red leather of the car door, even though this is a very dark shot.

The reduction in noise and increase in real sensitivity also helps the super slow motion. Compared to an FS7 I think the 120fps footage from the FX9 looks much better. It seems to be less coarse and less grainy. There is still some aliasing which is unavoidable if you scan the sensor at a lower resolution, but it all looks much better controlled than similar material from an FS7.

DYNAMIC RANGE.

And when there is more light the camera handles this very well too.  At 1:07 you can see how well S-Cinetone deals with a very high contrast scene. There are lots of details in the shadows and even though the highlights on the boats are clipped, the way the camera reaches the end of it’s range is very nice and it doesn’t look nasty, it just looks very bright, which it was.

For me the big take-away from this simple shoot was just how easy it is to get good looking images. There was no grading, no messing around trying to get nice skintones. The focus is precise and it doesn’t hunt.  The low noise and high sensitivity means you can get good looking shots in most situations. I’m really looking forward to getting my own FX9 as it’s going to make life just that little bit easier for many of my more adventurous shoots.

For more information on the PXW-FX9 click here. 

Or take a look at the Sony website.

Sony Venice – A close look at the dynamic range and noise.

With Sony Venice X-OCN files to download!

I have been working with Sony’s colour science guru Pablo at the Digital Motion Picture Center at Pinewood, looking at the outer limits of what Sony’s Venice camera can do. A large part of the reason for this is that Pablo is developing some really nice LUT’s for use on dailies or even as a grade starting point (Pablo tells me the LUT’s are finished but he is waiting for approvals and feedback from Japan).

As part of this process we have shot test footage with the Venice camera for ourselves and also looked long and hard at test shots done by other cinematographers. Last week we were able to preview a beta version of the cameras dual ISO modes. This beta firmware allowed us to shoot tests at both 500 ISO and 2500 ISO and the results of both are equally impressive.

I can’t share any of the test footage shot at 2500 ISO at this stage. The firmware is still in it’s early stages and the final version may well perform a little differently (probably better). But I can share some of the footage shot at 500 ISO.

Please remember what we were exploring was the extreme ends of the exposure range. So our little test set was set up with some challenges for the camera rather than trying to make a pretty picture.

We have deep, deep shadows on the right behind the couch and we also have strong highlights coming off the guitar, the film can on the shelves and from the practical lamp in the background. The reds of the cushion on the couch look very different with most Rec-709 cameras as the colors are outside the Rec-709 gamut.

Another aspect of the test was to check the exposure rating. For this I used my Sekonic lightmeter to measure both the incident light and the light reflected by the Kodak grey card. My light meter gave me T4 at 1/48th for 500 ISO and this turned out to be pretty much spot on with what the scopes told us. So straight away we were able to establish that the 500 ISO exposure rating appears to be correct. We also found that when we stopped down by 2.3 stops we got the correct exposure at 2500 ISO, so that too appears to be correctly rated.

Once the base exposure was established we shot at 2 stops over and 2 stops under, so from T2 down to T8 using a Sony 35mm PL prime. We used the XOCN-ST codec as we felt this will be the most widely used codec.  When looking at the files do remember that the 16 bit XOCN-ST files are smaller than 10 bit ProResHQ. So these are files that are very easy to manage. There is the option to go up in quality to Sony’s linear raw codec or down to X-OCN LT. XOCN-ST sits in the middle and offers a nice balance between file size and image quality, it being very hard to find any visual difference between this and the larger raw files.

The files I’m providing here are single X-OCN frames. They have not been adjusted in any way, they are just as shot (including being perhaps a touch out of focus). You can view them using the latest version of Sony’s raw viewer software or the latest version of DaVinci Resolve. For the best quality preview, at this time I recommend using Sony’s Raw Viewer to view the clips.

Click here to download these Venice Samples

If you find these files useful please consider buying me a coffee or beer.


Type



pixel Sony Venice - A close look at the dynamic range and noise.

So what do the files look like? First I recommend you download and play with them for yourself. Anything I do has to have a LUT,  grade or other process applied so that the linear data can be viewed on a normal computer screen. So it’s better to take a look at the original files and see what you can do with them rather than just accepting my word. The images here were create in DaVinci Resolve using ACES. ACES adds a film type highlight roll-off and uses film type levels, so the images look a touch dark as there were a lot of low light level areas in the test shots.

Venice at T4 The base exposure for the test.

Venice-base-T4_1.1.1-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T4 (From ACES). This was the “base” exposure for this test. Click on the image to enlarge.

Venice at T8 – 2 Stops under exposed (As exposed).

Venice-T8-uncor_1.4.1-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T8 (2 stops under). Click on the image to enlarge.

Venice at T8 – 2 Stops under exposed (Brightness corrected to match base exposure).

Venice-T8-norm_1.4.2-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T8 (2 stops under). Brightness match to base exposure via metadata shift. Click on the image to enlarge.

Venice at T5.6 – 1 stop under exposed (brightness corrected to match base exposure).

Venice-T5.6-norm_1.5.1-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T5.6 (1 stops under). Brightness match to base exposure via metadata shift. Click on the image to enlarge.

Venice at T4 The base exposure for the test.

Venice-base-T4_1.1.1-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T4 (From ACES). This was the “base” exposure for this test. Click on the image to enlarge.

Venice at T2.8 – 1 stop over exposed (brightness adjusted to match base exposure).

Venice-t2.8-norm_1.2.1-1-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T2.8 (1 stops over). Brightness match to base exposure via metadata shift. Click on the image to enlarge.

Venice at T2.0 – 2 stops over exposed (brightness adjusted to match base exposure).

Venice-T2-norm_1.3.2-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T2 (2 stops over). Brightness match to base exposure via metadata shift. Click on the image to enlarge.

Venice at T2.0 – 2 stops over exposed (as shot).

Venice-T2-uncor_1.3.1-1024x540 Sony Venice - A close look at the dynamic range and noise.
Venice at T2.0, 2 stops over, as shot. Click on the image to enlarge.

NOTES AND OBSERVATIONS:

I shouldn’t rush these tests! I should have set the focus at T2, not at T4. Focus is on the chart, not the dummy head. It would have been better if the eyes and chart were at the same distance.

It’s amazing how similar all the shots across this 5 stop range look. Just by adjusting the metadata ISO rating in Resolve I was able to get a near perfect match. There is more noise in the under exposed images and less in the over exposed images, that’s expected. But even the 2 under images are still pretty nice.

NOISE:

What noise there is, is very fine in structure. Noise is pretty even across each of the R, G and B channels so there won’t be a big noise change if skewing the white balance towards blue as can happen with some other cameras where the blur channel is noisier than red or green. Even at T8 and 2 stops under the noise is not unacceptable. A touch of post production NR would clean this up nicely. So shooting at 500 ISO base and rating the camera at 2000 EI would be useable if needed, or perhaps to deliberately add some grain. However instead of shooting at 500 ISO / 2000 EI you might be better off using the upper 2500 base ISO instead for low light shoots because that will give a nice sensitivity increase with no change to the dynamic range and only a touch (and it really is just a touch) more noise.

If shooting something super bright or with lot and lots of very important highlights  I would not be concerned about rating the camera at 1000EI.  For most projects I would probably rate the camera at 500EI. If the scene is generally dark I may choose 400EI just to be a touch cleaner. With such a clean image and so much dynamic range you really can pick and choose how you wish to rate the camera.

Venice has more dynamic range than an F55 and maybe a bit more than the F65. Most of the extra dynamic range is in the shadows. There is an amazing amount of picture information that can be pulled out of the darker parts of the images. The very low noise floor is a big help here. In the example below I have taken the base exposure sample and brought the metadata ISO up to 2000 ISO. Then I have used a luma curve to pull up the shadows still further. If you look at the shelves on the left, even in the deep shadow areas it’s possible to see the grain effect on the dark wood panels. In addition you can see both the white and black text on the back of the grey book on the bottom shelf. Yes, there is some noise but my meter put these areas at -6 stops, so being able to pull out so much detail from these areas is really impressive.

Venice-deep-shadows_1.1.2-1024x540 Sony Venice - A close look at the dynamic range and noise.
An amazing amount of information still exists in even the darkest shadow areas. This image adjusted up significantly from base exposure (at least +4 stops).

In the highlights the way the camera reaches it’s upper limit is very pleasing, it does seem to have a tiny roll off just before it clips and this looks really nice. If you look at the light bulbs in this test, at the base exposure, if you bring the highlights down in post you can see that not all of the bulb is completely over exposed they are only over exposed where the element is. Also the highlights reflecting off the guitar and film can on the shelf look very “real” and don’t have that hard clipped look that specular highlights on other cameras can sometimes have.

Another thing that is very nice is the colour tracking. As you go up and down in exposure there are no obvious colour shifts. It’s one of the things that really helps make it so easy to make all 5 exposures look the same.

The start up time of the Venice camera is very impressive at around 6 to 8 seconds. It’s up and running very quickly. The one stop steps in the ND filter system are fantastic. The camera is very simple to use and the menu seems logically laid out. It’s surprisingly small, it’s not much bigger than a PMW-F55, just a little taller and a little longer. Battery consumption is lower than most of the competition, the camera appears to consume around 50w which is half the power consumption of a lot of the competition. It can be run of either 12v or 24v. So all in all it can be rigged as a very compact camera with a standard V-Lock battery on the back.

Looking forward to shooting more with Venice in the very near future.