How Good is the Panasonic GH5? Lab Review + Free GH5 LUT [UPDATED]

March 1st, 2017
How Good is the Panasonic GH5? Lab Review + Free GH5 LUT [UPDATED]

 [UPDATED:] This article has been updated. Read the update notes and clarification HERE.

Panasonic GH5 Review - Sensor

The Panasonic GH5 is one of the most highly-anticipated compact cinema cameras of the year. It is the first of its kind to finally break the barrier towards a compact form-factor 10-bit 4K acquisition in the realm of (somewhat) large sensor cameras. Here is our lab analysis, where we will evaluate all the pros and cons of the new GH5 sensor, with a focus on the new 4K 10-bit mode.

Panasonic GH5 Review

After a short delay, here is our long awaited Panasonic GH5 lab review. Also check out the preliminary hands-on that Graham did in January.

Even though we try very hard to give you an objective view of the product reviewed, please note that all information and conclusions written in this article reflect the opinion of the author. We recommend that you always try and test any product for yourself and find your own conclusions based on your technical requirements before you use them.

Panasonic GH5 with a Speedboster Ultra and the Zeiss 50mm Cp2 Macro Lens

Panasonic GH5 with a Speedboster Ultra and the Zeiss 50mm Cp2 Macro Lens

In this Panasonic GH5 review, I will run the usual tests I do here at cinema5D, and invite you to check several important sensor attributes like dynamic range, sharpness and rolling shutter performance. I will also try to examine the low-light performance and take a particular look at 10-bit.

We test all cameras with the same Zeiss 50mm Cp2 macro lens (more on how we test HERE). In the case of the Panasonic GH5, the lens was attached via EF-mount with the Metabones Speedbooster Ultra adapter, which in my experience gives me excellent optical performance. For the dynamic range test I use a DSC labs XYLA-21 transmissive test chart, and to measure sharpness I used a combination of a subjective chart and a huge 8K Imatest ISO chart. Let’s get started.

The Panasonic GH5 Introduces 10-bit 4K

That the Panasonic GH5 introduces 10-bit color depth is this camera’s most interesting feature, and one that had many filmmakers (including myself) excited for the last couple of months. But looking at the camera’s specs, there are other attributes to like:

  • Slightly larger sensor area than the GH4
  • 5-axis internal image stabilisation
  • Internal 10-bit 4:2:2 in 4K (up 30fps)
  • Internal 8-bit 4:2:0 in 4K (up 60fps)
  • Up 180fps in 1080p
  • Dual SD card slots
  • No recording time limit
  • V-Log Gamma (paid upgrade $100)
  • Affordable price

A firmware update later this year will bring 6K photo mode, anamorphic recording as well as internal DCI 4K 10-bit with a 400mbit all-I codec. Currently, we’re stuck with 150mbit, which is fine too and should give us great results.

But let’s go into 10-bit color depth right away. I was really eager to test it and expected to see silky smooth skin gradations and footage with rich color information from Vlog recordings. After all, many of us have been shooting 8-bit color on most of their productions and the step up to 10-bit should make a huge difference, going from millions to 1.1 billion colors of file information. Furthermore, many were frustrated with Vlog on the GH4, so let’s try to see how the performance was improved on the new model.

[UPDATED:] Panasonic GH5 10-bit & Vlog

I was impressed when I first saw GH5 10-bit footage. With the right coloring or LUT, the results looked more organic and filmic than most of what I had seen in the area of small form-factor cinema cameras that shoot 8-bit. But after trying Vlog and tweaking the colors a bit more, I also found that skin tones were not as realistic as on larger cinema cameras like the Panasonic Varicams or cameras like the Canon C300. That said, I found that the 10-bit images from the GH5 still gave me much more control over my colors than other small form factor 8-bit cameras available today.

To get a better idea of how good the 10-bit quality of the Panasonic GH5 really is, I did a subjective side-by-side comparison with a few other cameras in a post production workflow suitable to the camera’s price tag.

Workflow

At the time of my review, 10-bit files from the Panasonic GH5 were incompatible with Adobe Premiere Pro, After Effects and Blackmagic Design DaVinci Resolve, which is the software that I use on a Mac Pro, 2013 for video editing. The only way to get footage into the editing software was to first transcode it in a different app. I used Apple’s Final Cut Pro X.

In an effort to make my tests as “real world” as possible, I chose the familiar Apple ProRes HQ codec, not only because it is widely used by professionals in 10-bit environments and external high quality recordings, but also because it was used as the recording format of the Canon C700 that I compared the Panasonic GH5 to and, because it is described by Apple as a codec that preserves “… visual quality at the same high level as Apple ProRes 4444 but for 4:2:2 image sources.” and “… supports full-width, 4:2:2 video sources at 10-bit pixel depths, while remaining visually lossless through many generations of decoding and reencoding.”.

Adobe Premiere Pro & Apple ProRes HQ Import Error

The consequence of using this codec however led to incorrect results when evaluating the Panasonic GH5’s 10-bit Vlog performance as it affected my files in a way that introduced an unexpectedly large quality loss within Adobe Premiere Pro. It seems like the quality issues previously found in my original test were caused by an error of the Adobe Premiere Pro software in connection with the Apple ProRes HQ codec. While the reasons for the error have not been clarified, anybody who intends to use Adobe Premiere Pro in connection with 10-bit files in Vlog recorded on the Panasonic GH5, should consider the conversion steps carefully. Using the Apple ProRes HQ codec on a daily basis, I was surprised that the conversion affected the Vlog files from the Panasonic GH5 in my initial test so severely. After reevaluation, it has become clear that the Apple ProRes HQ codec is more than capable to hold the 10-bit information Vlog requires and the error in connection with Adobe Premiere Pro needs to be reported and investigated. Furthermore the Panasonic GH5 in Vlog mode exhibits a magenta / green color shift in some parts of the image, that was further exaggerated by said error.

Panasonic told us they have been studying the color shift symptom in V-Log, including the compatibility with each of the color grading apps. And that they will make an announcement on their web site when they release an updated firmware to improve performance.

The exaggerated color shift was also visible during recording on the camera itself, in preview mode in Final Cut X, as was also present on native footage from the Panasonic GH4. However in 10 bit mode on the GH5, the color shift is much less severe and can be controlled much better than on the GH4.

Comparing the cameras

We had a Sony FS7, Canon C700 and Canon 1DC available at cinema5D HQ. Two of those cameras shoot 10-bit internal, and one of them is in fact a seriously expensive and high end camera (C700), so that should give me some perspective.

Panasonic GH5 Skin Tones Vlog – Minimal Grading, 10bit

I recorded my hand and face to see how the sensor would interpret skin tones. The color depth was set to 10bit, I shot in V-log and did a rough grade to match the other cameras. I tried to only work on curves, but had to reduce the overly pinkish red tones in order to keep the skin tones natural.

The window on the left is a 150% crop of my cheek with a 25% boost in saturation, so you can see the color separations more clearly. The image from the Panasonic GH5, when treated correctly seemed to have  a good amount of color depth to work with for me, even though the colors were not ideal right out of the box.

Panasonic GH5 Skin Tones - Minimal Grading, 8bit

Panasonic GH5 Skin Tones Vlog – Minimal Grading, 8bit

Here is the same shot in 8-bit mode on the Panasonic GH5. Color gradations seem much poorer and remind me of images I’ve seen shot with the Panasonic GH4.

It is true that a flat gamma like V-log greatly diminishes color information in exchange for extended dynamic range and softer highlights and shadows.  Clearly the 10-bit codec is the one to choose when shooting Vlog on the Panasonic GH5.

Canon 1D C Skin Tones - Minimal Grading, 8bit

Canon 1D C Skin Tones – Minimal Grading, 8bit

Here’s the Canon 1D C, shot in C Log Gamma. I can see a similar kind of reduced color information there as on the GH5’s 8-bit recordings that I did. The resulting images have large areas with a single color. However the overall skin color seems more harmonious and natural than on the GH5. Anyway, the 1D C is an outdated 8-bit camera for many, so let’s move on to other 10-bit cameras.

Sony FS7 Skin Tones - Minimal Grading, 10bit

Sony FS7 Skin Tones – Minimal Grading, 10bit

Here is the Sony FS7 with an Slog-2, 10bit recording in XAVC-I and there is certainly smoother color, almost comparable to that of the GH5’s 10-bit files, with skin tones that I find just “ok”. I could still make out some color artifacts in the skin, and bit depth seems to be a bit less than I had expected. Color depth of the GH5 seemed a bit nicer to me.

Canon C700 Skin Tones - Minimal Grading, 10bit

Canon C700 Skin Tones – Minimal Grading, 10bit

Here is where it gets interesting. The Canon C700 is a camera that comes in a huge form factor and with a high price tag. It is a serious moviemaking camera and the latest flagship Canon C model that also shoots 10-bit ProRes HQ in 4K DCI, so that’s what I chose for this shot. To me, this one gave me the best skin tones, neutral and harmonious colors and even though recorded in ProRes HQ, only minimal color artifacts and a color depth similar to that of the GH5.

10-bit Test Conclusion

Panasonic GH5 Skin Tones 10bit compared to other cameras

The bit depth on the GH5 when shooting in Vlog at 10-bit seemed very good to me. While I didn’t like the skin tones so much and found a slight color shift (white surface shades sometimes exhibited a magenta / green tone for me), it still gave me a lot of color information that I could work with.

Here is how the 8-bit codec compared to the 10-bit codec of the Panasonic GH5 looked like in my tests. I graded this with the official Panasonic Vlog to rec709 LUT and increased contrast:

Panasonic GH5 10 bit color depth.

Many people had hopes that the GH5 would solve many of the problems everyone had had with Vlog on the 8-bit GH4. To me it seems like those problems were addressed with the new 10bit 422 recording option. It looks like the 8 bit codec still suffers from a similar color shift as seen on the GH4.

Panasonic told us they have been studying the color shift symptom in V-Log, including the compatibility with each of the color grading apps. And that they will make an announcement on their web site when they release an updated firmware to improve performance.

8-bit / 10-bit Quality in PP “Neutral”

Looking at a different picture profiles I found that the color gradations looked nice even in 8-bit mode, and nicer than on most other 8-bit cinema cameras I’ve seen before.

Panasonic GH5 – Picture Profile: “Natural” – Nice color gradations

As you can see in the shot above I used PP “Natural”. The results were pleasing, both in 10-bit as well as in 8-bit mode, when converted using a high quality codec like Apple ProRes 4444. Even when I pushed the image’s contrast I saw that there was a lot of room for grading before the image finally fell apart:

Panasonic GH5 – Picture Profile: “Natural” – Contrast Pushed

The difference between the 10-bit recording and the 8-bit recording in a picture profile other than Vlog seemed minimal. When I pushed the contrast to the extreme (image above) I could spot where the colors break apart, but they did so very similarly on both recordings. 8-bit mode looked good to me and if you shoot in any codec other than 10 bit, I would actually recommend to avoid Vlog for now. In 10bit however, Vlog is certainly recommended.

Dynamic Range

Now let’s focus on another important camera attribute – dynamic range. A good dynamic range rating allows us to capture a larger range of shadows and highlights in high-contrast scenes – an important property when it comes to evaluating a camera intended for filmic use.

Our software was very strict with the GH5. We always take noise into account, because in all honesty, a very noisy image can only be recovered up to a limited degree as image information is quickly lost. With our usual SNR threshold of 1/0.5, our software measured a rough 10 stops of usable dynamic range on the Panasonic GH5. This is a slightly sub-normal rating. Both the Sony a7S II and the Fujifilm X-T2 have a higher rating. Here’s a screenshot of the dynamic range of a few popular cameras:

Panasonic GH5 Dynamic Range Test

Panasonic GH5 Dynamic Range Test

We can see that the sensor’s total dynamic range is higher than those 10 usable stops, but in a real world scenario I think the noisy shadows will have to be cut away in grading. When used correctly, this camera can still produce admirable results. On a subjective scale – and if you don’t mind some noise in your images – I think this camera can give you 11 usable stops when you’re careful. You should just try to avoid underexposure on the Panasonic GH5.

[Update:] Many people suggested that they would be able to use noise reduction in order to get back more image information. While I personally find this unrealistic, it might be interesting for those people to know, that the total patch range detected by the software (regardless of considerable noise) was about 18 stops, the same as on the Sony a7S II, while it is 20 on the Arri Alexa. I hope this information is helpful for those who prefer to draw their own conclusions in this segment.

Noise and Low Light

In terms of noise and low light, I found that the camera actually doesn’t perform as bad as many feared (the GH4 had rather limited lowlight capabilities). In my tests the camera actually seemed to produce those same 10 stops of dynamic range at both ISO 400 and ISO 800. Quality only slightly degenerated at ISO 1600, and was at its breaking point at ISO 3200. At ISO 6400 the image clearly has much more noise and the color in the dark areas shifts towards pink and green. I think I would only shoot beyond ISO 3200 with caution.

In this image you can see a 100% crop of dynamic range steps 6 to 14:

Panasonic GH5 Lowlight ISO 800 to ISO 6400

Panasonic GH5 Lowlight ISO 800 to ISO 6400

On top of that, the MFT sized sensor and MFT mount also allows us to use the Metabones Speedbooster Ultra, which gives us an extra stop of exposure range and nice super35 shallow depth of field. So even though the out-of-thebox lowlight performance is only average, the Speedbooster can help us out a bit and for me, improves the overall look as well.

Panasonic GH5 and Metabones Speedbooster Ultra

Panasonic GH5 and Metabones Speedbooster Ultra

Rolling Shutter

The Sony a7S II suffered from strong rolling shutter problems, a phenomenon also referred to as “jello”. The rolling shutter phenomenon that we see on most CMOS sensor video cameras is also present on the Panasonic GH5, but is much less severe in comparison to many other cameras.

Panasonic GH5 Rolling Shutter

In my experience, after having tested the rolling shutter of over 50 different cameras, smaller sensors usually give us better rolling shutter ratings and the Panasonic with its MFT sensor seems to follow that pattern as well.

I used a rotating chart and measured the horizontal distortion of a vertical line to evaluate the total delay of sensor line readout.

Image Quality

Image Detail is the aspect where the Panasonic GH5 performs very well. To me the overall image looked organic with very nice detail, comparable to cameras like the Fujifilm X-T2 or FS7.

Panasonic GH5 Detail

100% crops for Image Detail

In terms of aliasing, the Panasonic GH5 also seemed to perform very well. In comparison to the FS7 though, I thought it was a bit weaker. “Horizontal resolving power” could be observed on this crop of our 8K test chart where lines couldn’t be separated as nicely as on the FS7.

Panasonic GH5 Vertical Aliasing

Panasonic GH5 Vertical Aliasing

Panasonic GH5 vs. Sony a7S II - Subchart Image Details

Panasonic GH5 vs. Sony a7S II – Subchart Image Details

The Sony a7S II has a problem with contrast areas. This is where the camera’s image gets soft and loses detail for unexplained reasons. The Panasonic GH5, in comparison, seemed to resolve all aspects of image detail quite nicely and seemed to outperformed the a7s II in that regard. I think this becomes very clear in the shot above. The slight grain (noise) on the Panasonic GH5 looked nice and filmic to me.

When using Vlog, I found it was important to set Sharpness (S) and Noise Reduction (NR) to -5. This gave me the best quality on the Panasonic GH5. Otherwise the image would look oversharpened to me and noise reduction didn’t control noise in any significant way for me when set to 0.

180fps Slow Motion

It is nice that the Panasonic GH5 gives us the option to shoot 180fps in HD resolution in such a small form factor. However, as on most other cameras this feature seems much better on paper than in reality, because the true resolution and quality seemed to be much less than 1080p in my tests.

Panasonic GH5 Slow Motion 180fps Image Quality compared to 25fps.

Panasonic GH5 Slow Motion 180fps Image Quality compared to 25fps. 100% crop.

As you can see the quality in HD resolution was very nice in my tests, whereas in slow motion mode, at 180fps, the image had visibly lower resolution and leaned more towards an actual 720p. Also I found the image exhibited the magenta / green color shift when used with Vlog Gamma, so I would only use slow motion with one of the other picture profiles.

Conclusion

This has become a very long winded article and countless hours we spent checking, testing and retesting, but some of the issues I found required some investigation and explanation and I hope you found these findings useful. In conclusion I think it is no doubt that the Panasonic GH5 is a very tempting offer when you look for an affordable, compact cinema camera that can potentially produce some great looking footage. There is a 10-bit option, 4K has nice details and 180fps slow motion is also there. In practice I found that some of those features are a bit less convincing than others, but overall the GH5 is a camera that is an intriguing allrounder that many will choose.

Log gamma has become the goto option for filmmakers who want high quality cinematic results and avoid the “digital look” that some of those cameras can produce. Also it offers the most flexibility in grading and opens up a world of possibilities in terms of LUTs and camera matching. Fortunately in my tests it eventually turned out that Vlog seems very capable with the GH5’s 10-bit mode. Even though I was not fond of the skin tones and the color shift, it seemed like the bit depth of the camera can actually match that of other and more expensive 10-bit cameras out there and gives you a lot of room in grading. This is great news for any filmmaker on a budget, because this camera has a lot in store.

If you are happy with the colors of the GH5’s Vlog, then I think the GH5 is currently one of the most convincing offers as an affordable cinema camera. Even though most of its sensor performance (detail, dynamic range, lowlight, 180fps) was only average to me, there are several aspects to enjoy too: The internal image stabilization was a pleasure to use for handheld applications, it gave me a very cinematic handheld look, the peaking is very useful and special features like the histogram and anamorphic recording options make this a more cinema capable camera than most other small handheld devices before. The introduction of 10-bit in this form factor is certainly a game changer and I hope we’ll see more cameras like these pop up in the near future. Until then, the GH5 is definitely on our top recommendations list.

All in all, if you had your money set on 10-bit, the Panasonic GH5 certainly seems like a convincing option, doing away with most of what people didn’t like about the GH4 and presenting a great compromise between Blackmagic’s and Sony’s offerings at a very competitive price.

Pro’s

  • Good 10-bit 4:2:2 color depth in comparison to other 10-bit cameras
  • MFT sensor works very well with Metabones Speedbooster Ultra for nice shallow depth of field.
  • 5-axis internal image stabilisation
  • Advanced video shooting functions like peaking, histogram or anamorphic mode
  • Very nice quality 8-bit mode! Up to 60fps in 4K
  • Good image detail
  • Up to 180fps (Quality is not up to 1080p standards though)
  • Good rolling shutter perfomance
  • Dual SD card slots
  • No recording time limit
  • Affordable price

Con’s

  • reddish Skin tones and sometimes magenta / green color shift, especially in 8 bit.
  • Dynamic range is limited
  • Lowlight is average
  • 180fps mode has less quality than actual 1080p
  • MFT mount has limimtations (Some Canon lenses did not work with the Metabones Speedbooster)
  • MFT sensor is smaller than industry standard super35 (But gives you a great price and can be improved with the Metabones Speedbooster Adapter)

cinema5D instaLUT B1010 applied to Panasonic GH5 10bit

Panasonic GH5 LUT – cinema5D instaLUT FREE

This LUT is made for VLOG ON THE PANASONIC GH5 only! Vlog is a paid upgrade on this camera.

This LUT is ideal for people who don’t want to tamper around with colors too much or who want to get a nice and filmic basic look in post. This LUT imitates our instaLUT B1010 that we first used on the Arri ALEXA and DJI Inspire 2 here.

Download:

[ulp id=”URZ1kV2trvwr16Wo”]

To get our Panasonic GH5 LUT we ask only that you subscribe to our newsletter and in turn you will get the download link sent to your e-mail. This is a double opt-in and will not work with fake e-mails.
Your e-mail address stays with us. No spam or third parties and you can of course unsubscribe if you don’t like the weekly newsletter whenever you receive one. We hope you will enjoy our future content and stay on board.

Here are a few shots where you can see the LUT applied to Panasonic GH5 footage. These shots have the look applied with no other modifications.

250
Leave a reply

1
1
- 8
Filter:
all
Sort by:
latest
 Jeffrey Baker
Member
March 11th, 2017

Hi, I’m just wondering if you have heard anything back from Panasonic now that we are 10 days out from this post. I’m also interested in how this cameras autofocus compares to the a6500? Has anyone had a chance to test the autofocus capabilities for us run and gun shooters? Thanks!

Wolfgang Ernst
Wolfgang Ernst
Member
March 11th, 2017

Someone who’s leaning too far out of the window (“Lab Review Reveals 10bit Flaws in Vlog”) is in danger tipping over. As in good practice of journalism – placing a hot headline is half the rent. The journalist must aim to provide correct proof. Incorrect information must be corrected immediately, e.g. mixing codec compression errors with bit depth.

 Peter Smith
Member
March 10th, 2017

wow! great, in-depth review! It’s about time someone pointed out those issues we’ve also come to know from the GH4.

 Emil Ene
Member
March 9th, 2017

Dear Sebastian:

Thank you for your time to review some aspects of the upcoming Panasonic GH5.
Since I’m eagerly waiting for the release of this camera, and have read many reviews so far, I was looking hear your experience with this camera.

As I read your findings of course that raised questions to me concerning the aspects you mention, and specifically how they might relate to the real world usage for my intended use.

Some things I’m still very interested to read about is a look at the real world usage using the inbuilt picture profiles.

As I have been reading through the comments on the internet relating to the findings of your review, I have noticed some reasonably arguments by professionals concerning the this test and the interpretation of the results.

The no. 1 argument that stands out to me is, that an external recorder should be used to repeat the test, and the subsequent evaluation of footage. The result will be different they say, not based on their assumptions, but on prior similar experience, for example with GH4 Vlog and internal/external recording results.

Since this is not a matter of subjectivity, but something that can be easily objectively evaluated, why not buy/lend an external recorder, such as the Atomos Shogun Flame, and repeat the test?

Since, at the end of the day, truth will only become clearer when closely examined.

If these people are right, we have all learned something, and henceforth future reviews can be improved by also using an external recorder as part of the testing method. On the other hand, if it doesn’t solve the issue under discussion, the argument is proven incorrect, and still we all have learned something.

My advise is to look at the comments for any possible factual truth it may contain, in a positive way, forgetting the perhaps impolite manner in which it may be stated at times. And try to test and research the claims, as to determine what is actually true.

After you have established the accuracy/inaccuracy of an argument, you will help us all (including the critics) by releasing the information that will point out the facts.

I hope you will be encouraged to take up this challenge.
I’m sure we can all learn much from it!

Thank you again for the hours invested so far.

Regars, and all the best,

Emil

P.S. I will post here some comments I’ve found on the internet, that might be useful:

“Have you ever looked at any GH4 V-log material recorded externally? Plenty of people have and can see just how much damage the H264 does.”

“Plenty of examples of V-log on the GH4 recorded to ProRes that eliminate most of the color issues.

“Did anybody on here honestly expect 150mbps 4:2:2 10bit H264 to look as good as external ProRes422 HQ? While the color space enhancements are there with internal 10bit the encoder is still a H264 based encoder that needs to group similar colors together to get the video down to only 150mbps.

To 99% of the GH5 users the internal 10bit format will be phenomenal and they will fall in love with it. If a colorist will be touching any of the material however then I have always highly recommended using an external recorder and a format like ProRes422 HQ.

[…] Of course the internal format isn’t visually perfect with V-log. It is still a extremely highly compressed format using the estimation methods of an H264 encoder. You can optimize that until you are blue in the face but that is still 1/5th the size of ProRes422 HQ and even that is highly compressed compared to uncompressed. That 150mbps is 1/27th the size of UHD uncompressed. That is a lot of tossed out data. The fact that it looks as good as it does at 1/27th the size is exceptionally impressive. Of course there are color artifacts there. If you want the minimal amount of V-log color artifacts you record ProRes422 HQ which is a bit under 1/6th the size of uncompressed. While it may not be seen with the naked eye easily that extra data is there for grading.

It would have been nice to see Cinema5D actually test the GH5 with external recording since that opens up even more quality options. If a film maker is going to really get down and dirty with that level of pixel peeping quality concerns I would ask why on earth are they recording internal 150mbps when they have a perfectly good and affordably set of solutions for external ProRes422 HQ?”

“Yes internal 10bit will have some loss of data but I think 99% of the people will be hard pressed to actually notice it. The color issues are seriously splitting hairs and even then only really show up if you use V-log. The article even states the other profiles look very good. What is in question here is really: is H264 even at 10bits still not good enough for V-log? I think the answer is: it really depends. For most people who thought 4k on the GH4 looked amazing it will look even better on the GH5 and V-log is finally usable internal. With the Gh4 you could use V-log internally but not without some artifacts. With the Gh5 most of those are reduced to a point where they will not be noticed as much. So the 10bit internal is still a great format for those who really want V-log and don’t want to use an external recorder. We have seen plenty of great samples from film makers like Luke Neumann that show us just how good it can look. It is much better than internal V-log on the GH4, so a huge improvement.”

Thomas Smet

———————————————————

“Thomas is absolutely correct, and this test confirms what many have been saying from the start of the V-log L controversy: For 10-bit resolution, use an external recorder.

Inadequate bitrate has been a persistent pitfall with internal H.264 recording on Lumix cameras, ever since the GH1. There’s an inherent trade-off between SD card recording capacity versus image quality, and with Panasonic, mainstream consumer satisfaction always takes priority. Just as the GH4’s 100 Mbps 8-bit 4:2:0 4K bitrate was barely adequate, so is the GH5’s 150 Mbps 10-bit 4:2:2 4K bitrate. To capture those extra two bits of precision, the H.264 encoder has to preserve at least twice as much detail in both horizontal and vertical dimensions, making it far more susceptible to bitrate degradation from artifacts and low-level noise contamination. At 150 Mbps, most of those 10-bit details will inevitably be sacrificed in shots that involve significant motion or panning.

For comparison, I’ve shot a great deal of 8-bit 4:2:0 4K footage at 150 Mbps on the JVC LS300 in J-log1 mode. At that bitrate, the effective color bit-depth of this camera is excellent, but in extreme cases, 150 Mbps is barely enough to preserve smooth under-exposed gradients. For 10-bit resolution, it would definitely need more than 200 Mbps.”

“The reason Panasonic chose to use H.264 was because the GH line is a prosumer product […] V-log L and 10-bit 4:2:2 color depth were not anticipated in the original design, they were enhancements retrofitted into video capture formats that were already marginally inadequate for their original purposes. They will work fine for consumer-grade purposes, but their technical shortcomings will become visible when closely examined using high quality post-production tools. For optimal image quality, use of a high-bitrate external video recorder is essential.”

Lpowell

———————————————————

“Provide the original files for download, do another test with an external recorder, test the DR with the default NR settings (not minimum NR in camera).”

Eno

Member
March 9th, 2017

Until Sony releases the A9 at NAB next month ;)

 Boreko Treffertone
Boreko Treffertone
Member
March 8th, 2017

I´m sorry reading this test makes me request the competence of the author. Director and dop who studied at Filmacademy Vienna?
If someone with this education writes “It seems that none of these cameras truly records 1.1 billion colors and, unless my skin is made up of distinct color patches,…“ At what film school did´t they taught dept of colors is always linear and skin tone is at about 36%. By the way to improve this a log curve is made for. And if you want to show the difference between 8 bit and 10 bit you have to grade your picture heavily. Like this example made by slashcam.de and please don´t tell me you don´t see the difference.comment image

 Mike Hannon
Member
March 8th, 2017

https://www.slashcam.de/artikel/Test/Panasonic-GH5—8-Bit–10-Bit–4-2-2–V-Log-L—alles-OK—-Die-GH5-in-der-10-Bit-Praxis.html#Die_G

Slashcam’s test of internal GH5 10 bit versus internal GH5 8 bit do not appear to replicate the Cinema5D findings. It’s worth bearing in mind that they used a 32 bit working bit depth in After Effects for this test. I know that Final Cut is supposed to be 32 bit float as well, but is it possible that it’s not interpreting the 10 bit GH5 files correctly and that this accounts for the similarity of 8 and 10 bit in the Cinema 5D test?

Sebastian, can you examine your files in After Effects? (double click the 8bit icon in project panel to change to 32 bit)

Google translate link:
https://translate.google.com/translate?sl=auto&tl=en&js=y&prev=_t&hl=en&ie=UTF-8&u=https%3A%2F%2Fwww.slashcam.de%2Fartikel%2FTest%2FPanasonic-GH5—8-Bit–10-Bit–4-2-2–V-Log-L—alles-OK—-Die-GH5-in-der-10-Bit-Praxis.html%23Die_G&edit-text=&act=url

Member
March 7th, 2017

Well 2 things : when canon released the C300 mk2 Clog2 was rather noisy in the shadows. Clog2 was not the most usable setting, especially if you underexposed at all.

With Firmware 1.03 they changed how the read out the sensor, and added cLog3. With cLog2 pre firmware 1.03 they were stretching out the shadows too much plus with less than ideal sensor processing it wasn’t good. So in this case a firmware update that was probably significant under the hood in what the camera does led to more usable DR. Shadows are much improved with much lower noise levels, and cLog3 doesn’t push the shadows too much. A touch of NR at a setting of 2 quiets the noise in the shadows from low to near nothing.

so a firmware update can make a big difference IF THE HARDWARE is up up to the task. I’d suspect in the Gh5 there isn’t really that much extra the hardware can do. a higher bitrate recording isn’t that much of a push. better NR, or changing how you read out the sensor like canon did probably isn’t doable.

in reality, for the price, its a respectable camera. its certainly interesting to compare it to some vastly more expensive cameras, but at some point, well, you get what you pay for. expecting this to be a Varicam LT mini is a bit unrealistic. If you need big camera performance, well, the answer is you need one of those big cameras. this is of course the reason I’ve given up on dslr’s for video, they are mostly and generally too much compromise, lack of real video camera form factor, and other usability things like TC to make them viable for a lot of work. if the GH5 works for your needs great, but don’t complain its not a variacam LT mini for 1/10th the price.

Joseph Moore
Member
March 8th, 2017
Reply to  Steve Oakley

Noise reduction is a very important consideration when evaluating the GH5. One of the ways in which it eeks-out improved performance over the GH5 is much more sophisticated NR. (Most notably, analyzing temporal, not just spacial, changes.) By testing with NR set to -5, the camera was hobbled, and as such any evaluation of the real world image would have had to have included NR in post. Had this been done, even by Cinema5D’s methodology, and extra stop of DR would have arguably been revealed.

Member
March 8th, 2017
Reply to  Steve Oakley

Firmware yes absolutely. I think I already said when Panasonic released the DVX200 V-log was a shocking mess from a noise perspective ….. after the first firmware update the camera performed better in V-log then the GH4 using the same sensor.

The list of features (other then 10-bit) make this camera at 2k a no brainer, seriously. Hell even a full size HDMI Port …. LOL !!!!!

 m hazza
m hazza
Member
March 7th, 2017

i download the 422 10 bit 4k file from Neumann Films .. and i do Some color grading in adobe camera raw .. The results was very bad .. It does not look like 422 10bit .. the sky is falling apart after i Increase the Blue and change the luminance ..

 Charles Maring
Member
March 7th, 2017

Wow… That felt like a harsh review considering how much tech is packed into the GH5 for $2K! Technical tests are one thing, but in the real world I had a great experience with the GH5. Hands down my favorite camera to date. Personally I found the 10bit even in 150 to be slightly cleaner, and more forgiving visually, while working with a preproduction firmware in Standard mode. Anytime I tried to adjust density or push a grade with 8bit on any other camera the footage would fall apart much faster comparatively. The 10bit outperformed 8bit for sure in that regard, and in scenes with lots of color I could see a subtle difference. The difference isn’t astonishing, but I’ll take a more forgiving camera any day. I didn’t have an opportunity to test vLOG, but I suspect the firmware update with the dramatic bit depth increase will make the 10bit buttery smooth for those shooting feature films, etc… So, I would hold out on making any real solid conclusions until we see that.

That is just one small part of what I loved about the Lumix GH5 vs any other camera choice below $12k. I don’t have any budget concerns. In fact I have a closet full of Canon bodies and glass, but personally I find I can do a lot more with smaller mirrorless cameras vs larger cinema cameras or DSLRs. The 5 Axis is mind blowing for the work I do both professionally and personally, and the ergonomic on the GH5 were the best I have felt in a camera system with that built in.

Just my two cents from a very brief experience with the camera, and comparing it to the GH4. Here are a few thoughts for anyone interested in investigating the GH5 further. https://www.youtube.com/watch?v=K2pRNfPzrp4&t

 Illya Friedman
Illya Friedman
Member
March 7th, 2017

Cinema5D also claims 14.1 stops of dynamic range from the A7S, which it does not. It does not take an imaging scientist to tell you this. Regardless of any claims made here or by other bloggers, adding an extra light modifying optical element in between your lens and camera (in this case a “speed booster”) does not increase quality. This might be repeated by many bloggers as “fact” or “conventional wisdom,” but it is not. You should have very healthy skepticism of any claim or test result from someone who tells you that it does.

As a disclaimer- I say this as someone who has a day job selling all the products mentioned in my comment and in this post. I also have 20+ years of expert experience, including working as a technical consultant for some of the largest feature films and television series of all time.

Illya Friedman
Hot Rod Cameras

 Leonardo Fallucca
Leonardo Fallucca
Member
March 7th, 2017
Reply to  Illya Friedman

Actually they claim 12 stop of dynamic range for the Sony A7s, not 14. Your comment is not accurate.

Joseph Moore
Member
March 7th, 2017

Agreed. While I find their DR ratings to be just a tad conservative, they are applied even-handedly, and I’d rather they were conservative than unrealistic. “Usable dynamic range” is always going to be subject to interpretation.

 Illya Friedman
Illya Friedman
Member
March 8th, 2017

Sorry Leonardo, the chart may now be removed from Cinema5D, but you can still find a screenshot from when it was first posted over at NoFilmSchool. They list the Arri Alexa at 14.5 which is also inaccurate.

http://nofilmschool.com/2014/07/sony-a7s-dynamic-range-arri-alexa-amira

Joseph Moore
Member
March 8th, 2017
Reply to  Illya Friedman

To be fair, that was a couple of years ago using a different methodology, I believe.

 Illya Friedman
Illya Friedman
Member
March 8th, 2017

me too what?

 Kevin Winkler
Member
March 6th, 2017

I would love to see a comparison to 1dx mark ii

 John Norton
John Norton
Member
March 6th, 2017

Thanks for the review. “We were told firmware was pre-production (v1.00)” I think you should have waited for production firmware, even if it turns out to be the same as you tested. I also strongly believe that one of the cameras you used should have been the GH4, sure, no 10 bit to compare with, but plenty of other bits, dynamic range RS etc.

I noticed that the reasons for using the other cameras in the review became obvious only through answering some comments. Stating methology of testing, especially choice of other cameras used and the reasons for using specific cameras UP FRONT would make for more clarity, even if not everyone agrees with the other cameras chosen.

A conclusion with some pertinent TAKEAWAYS at the end of the review, something more comprehensive than the pros/cons might be an idea. After all, you went to such trouble doing all the testing, so why not flesh it out with a good quality comprehensive summary section, unless your heart just isn’t in the camera?

Like it or not, some people, in a hurry, just check out the conclusion first, then maybe when they have time later read the whole review, so I believe it’s important.

Joseph Moore
Member
March 6th, 2017

Much of my negativity to your article is due not to the results (which speak for themselves) but to the incomplete, erroneous, click-bait-ish conclusions made from them, and your unwillingness to correct, or at least qualify, them.

Indeed. V-LOG L (the correct name for the profile) is a compromised profile (a small curve designed to fit inside the larger Varicam V-LOG curve) that is being expressed in a compromised signal (either 8-bit 4:2:0 or 10-bit 4:2:2) and then recorded in a compromised codec (h264, low bit rate, designed for consumer delivery not acquisition.) When you then try to expand that back to full contrast, you emphasize the flaws introduced at each of the prior stages.

You could have taken some very simple steps – most obviously to introduce an external recorder into the mix – to verify the source of the weakness, and some equally obvious steps to highlight the improvements – most obviously to compare to the GH4. You did neither, and instead compared it to a competitors most expensive cinema camera.

We could quibble about some of your methodology (especially shooting resolution charts through an optical adapter!) but what I find most disheartening, is that by only focusing on charts, you’re leaving your readers with the false sense that there is something wrong with this camera. In fact, that is almost certainly not the case. At least not anything you’ve noted. (There does seem to be more noise in 10-bit recordings, for which there is no obvious reason, but that might have been cleared up with v1.0.)

The reality is that the GH5 improves upon the GH4 in every meaningful way. Image quality, stabilization, crop factor, ergonomics, menus, ports, etc., etc. ALL improved, based on customer feedback from the GH4.

I know that there is a desire to rush out information to hungry readers and to generate page views with a salacious “scoop”, but the responsible thing would have been to hold back until you could write a full review, and if you were really serious, to wait for Panasonic to have a chance to respond.

 Kevin Almodovar
Kevin Almodovar
Member
March 6th, 2017

Sebastian, you guys do great work and I love the reviews.

I think that sometimes you have to look at the bigger picture and understand that the GH line caters to a certain demographic that wants to know how much better the GH5 is to its predecessor.

There are by far, more people asking the question, “Is it worth it to sell my GH4 to get the GH5?” than there are people asking “Should I get a 1DC, FS7, C700 or a GH5?” I agree that you should always have at least one high end camera to see how much you can push a budget friendly tool, but as techies, we always want to know that manufacturers are improving their technology in equal to the new prices that they are charging. For example, how many people thought the C300M2 was way overpriced for what it brought to the table?

So to a Gh4 user this review leaves very little to be desired and maybe talking to those users before a test would give you a huge insight into what they want in a new camera and to see if Panasonic delivered.

 Kevin Almodovar
Kevin Almodovar
Member
March 5th, 2017

I think a better comparison for many of these specs would have been to the GH4. Especially since you use a different method form the manufacturers to determine dynamic range. Would be nice to know if it has improved over the GH4 and by how much.

Member
March 5th, 2017

It would be interesting to hear what you think about actually working with the camera, from a video-photographer point of view! How are all the menus, the feel of the camera, handling with speedbooster etc. All cred for other sites reviewing gear but I value your opinion since you know what you’re doing. :)

On another note, how many Mbit/s is the C700 and the FS7 recording their 10 bit with? IPB or ALL-I codecs?

Thanks for an interesting article!

 Matt Wolcott
Member
March 4th, 2017

Thank you for taking the time to create this review! I have the GH5 on pre-order and now I’m unsure of whether to cancel or not. It seemed too good to be true, and I just don’t know whether it is or not!

Of course I’d rather have the Ursa Mini Pro, but that’s 3x the cost. Not sure what I’ll do.

Member
March 4th, 2017
Reply to  Matt Wolcott

Matt, How does this look to you ?

https://www.youtube.com/watch?v=jA4ItA2iAK0

 Matt Wolcott
Member
March 5th, 2017
Reply to  Clayton Moore

That looks alright to me! What I wonder about is how something I shoot will hold up when I start grading. The 10-bit internal sounds so appealing for that reason, and I hope that it helps with banding issues. I will keep my order, and hope for the best.

Member
March 6th, 2017
Reply to  Matt Wolcott

And let us all know ?

Johnnie Behiri
Admin
March 4th, 2017
Reply to  Matt Wolcott

Matt.

Please don’t cancel your order. Best will be to get the camera, test it for yourself and see if it can serve you well.

Make sure you are buying it from a reputable place and can send it back in case you decide to do so.

Thank you.

Johnnie

Joseph Moore
Member
March 6th, 2017
Reply to  Matt Wolcott

There are so many *real* reasons to use this camera, and one review focused on one *potential* issue.

If you have to use V-LOG L, and if you do experience real world challenges, return the camera.

 marco porotti
marco porotti
Member
March 3rd, 2017

probably, just say probably, long GOP codec should give this kind of problems by its own nature (it save the information of one frame than just the variation between frames), ALL INTRA save every single frame (compressed but all frames) and, again, probably, differences in gradient could be better (in my fs7 I notice a lot of difference between all-I and long GOP).
After all this comments I wait a sample file recorded externally (just to bypass the codec).
That said thanks a lot for your good job and all the time that you spend to share your thoughts.
M.

Member
March 3rd, 2017

I believe a C5D should call a press conference with prime minister Sebastian Wöbe and Press Deputy Jhonnie Behri, so all these question can be addressed in a timely manner. Berlin is furious that your scientific perceptions maybe false, and Tokyo is concerned that their sales will lag thanks to this “somewhat damming report” There are questions in the Vlogsphere, skateboarders in LA are up in Arms, Washington is saying these are fake news and Brides all over the world may not get married because the camera has a fake 10 bit that makes them look ugly. Answers please.

Johnnie Behiri
Admin
March 3rd, 2017
Reply to  Vlad Box Rojas

VBR, We should hire you as a writer! Good description of the current reality…:)

Berlin took (again) the wrong pills. Tokyo is seriously investigating our findings. Washington is talking to Nevada and might forbid us to attend NAB and the brides are requesting to use the GH4 for now.

More questions please.

Have a peaceful weekend!

Member
March 3rd, 2017
Reply to  Johnnie Behiri

Hehehe…..You guys have a great week end.

Member
March 3rd, 2017
Reply to  Johnnie Behiri

In the end, its just a camera, its not NIRVANA, its not going to cure cancer. Buy it or not. If you want to see how it looks, watch for Johnnie Behiri to shoot some video with it.

There is one thing not in dispute Johnnies ability to make great video (whatever the camera). Even more, he is one of the most skilled story tellers you will see in video no longer then 2-3 minutes(Probably longer Ive just not seen one longer). But he is better then most Ive seen on line anywhere in that regard!

Johnnie Behiri
Admin
March 4th, 2017
Reply to  Clayton Moore

Clayton Moore

Thank you for putting things in perspective and also thank you for your kind warm words. Appreciated!

Luckily, many of my professional colleagues (also from other sites) are doing a great job testing and reviewing cameras.

Thank you again.

Johnnie

GELAX STUDIO
GELAX STUDIO
Member
March 3rd, 2017

Hey,guys,does the IBIS still work with Metabones adapter and canon lens?
I knew that the IBIS on sony cameras can also work with third party lens on ,don’t know if GH5 works the same.

Member
March 2nd, 2017

Very interesting. I was about to be pissed off that the specs of the GH5 match my Canon C300 MK II, but now I see that specs are specs and real world usage and testing are something else. Although I would like to get the GH5 for its small and powerful form factor…so far the C300 MKII and the XC10 are doing well together and my main go-to film making combo. I will defiantly keep my eye on this GH5 though! Good test…

Member
March 2nd, 2017

Oh Boy have we a whole bunch of NASA scientist here!

Member
March 3rd, 2017
Reply to  Vlad Box Rojas

Get on the CD5 couch with a few (not shooters) video engineers and have a good discussion, there are people who know a great deal about the technology. Shooters are shooters, not engineers.

Member
March 3rd, 2017
Reply to  Clayton Moore

Make new friends at the HPA Tech Retreat in Palm Springs or ask around at Pro Video Coalition there are people who shoot and who have engineering chops.

 Richard James Milne
Richard James Milne
Member
March 2nd, 2017

Wow… this is a PRE-PRODUCTION GH5! Coming from credible source as “Cinema5D” I would have not expected to do such a TEST on a PRE-PRODUCTION Camera, then bash it! Makes me want to take a look at who your sponsors are and unsubscribe!!!

Your credibility just in the toilet… Shame on YOU!!!

 Nicola Verdi
Nicola Verdi
Member
March 4th, 2017

Hi Sebastian,
Thank you for the review and first look. Great Job.


How old are you?

 Richard James Milne
Richard James Milne
Member
March 4th, 2017
Reply to  Nicola Verdi

Not that it’s your business to ask Nicola, however I am 60 years old. I have been a successful profession Photographer my entire life and have owned my own studio for 28 years, now enjoying retirement on Vancouver Island. I know a bit about cameras and reviews!

Are you a boy or girl?

 the SUBVERSIVE
the SUBVERSIVE
Member
March 2nd, 2017

It’s true that you can see macro blocking issues but is it a V-log Color Depth issue?

I mean this looks much more like compression issue and more importantly, how can you confirm it’s color depth related without even ruling compression out with an external recording sample before?

This whole flawed methodology makes me question if you guys really understand what you are doing or what you claim to know.

It would be nice if Panasonic could solve it via FW, but it might be a limitation of the 150Mbps IPB H.264 codec after all, which explains why a more robust AVC Ultra based codec from their Varicam division might be needed.

Hacks in the past have shown that tweaks and higher bitrate can help with image quality and compression issues, but I doubt that will happen. Either way, Panasonic could bump the bitrate to 240Mbps instead.

But talking about color depth, chroma subsampling, compression types, codecs, etc. this is already a complicated matter to discuss since few people actually understand it but then you go spreading misinformation?

Every couple of articles I see people in the comments pointing out technical and other mistakes in C5D articles, but this seriously damages any credibility one had in C5D. There is a macro blocking issue, but the way it’s written this is not only blown out of proportion but worst of all, the cause and possible solutions are misreported as well.

I personally dislike rushed reviews like this, no external recording tests, James Miller have been doing tests for a while and at least he did proper tests and even found out ways to work around the compression issues. For me, this doesn’t change my perspective either about the camera or about C5D, unfortunately, but it does for a lot of people that comes here seeking for accurate info about things they don’t understand or want to learn more about and this article is terrible in that regard.

 Greg Althoff
Greg Althoff
Member
March 2nd, 2017

I found this video today which makes me think this camera has more potential than first thought (excuse my other rant).
https://www.youtube.com/watch?v=duiWkHOLYoM

Time will tell..

Member
March 2nd, 2017

Wooa Twiter storm from Berlin is non stop. Anyway pixel peeping, not my thing, am into getting paid for good content and good looks. If I was to buy a camera this would not deter me from buying it. Good lights Good Temp Good Comp allows you to save money in post.

Johnnie Behiri
Admin
March 2nd, 2017
Reply to  Vlad Box Rojas

Yes, all we need is a sofa…

Member
March 2nd, 2017
Reply to  Johnnie Behiri

I hear you Johnny.

 LORDD LORDD
LORDD LORDD
Member
March 2nd, 2017

http://vimeo.com/205130607

Another test using the camera in VLOG and baking in a LUT directly into the Atomos Shogun. Really love how this works together and so well. Interestingly it produces a superior result over adding the same LUT in post. macro blocking and picture breakup are very much reduced.
Shot in VLOG partly in 4k UHD 25p and 4k UHD 25p downscaled in camera to 1080p to test for any induced aliasing etc.
Lens used was the Zeiss Otus 55mm and the MFT to EF Metabones Speedbooster Ultra x .71
Graded with DELUTS VLOG set (update for the GH5 coming soon)
Spot the one shot on the 135mm with the stabiliser set to 39mm.

Member
March 2nd, 2017

Excellent article Sebastian. I’ve been really curious (as are many others) as to how V-Log handles with the added internal 10bit 4:2:2 recording. Dang, what a bummer that the magenta blotches look exactly the same as on the GH4. Guess we’ll still have to resort to Atomos recorders to get around that issue. Speaking of which, do you have any plans to update this post with ProRes recording results? I think many people will record this way, especially since Panasonic added a full sized HDMI port.

P.S. Thanks for the LUT!

 Dan Edmondson
Dan Edmondson
Member
March 2nd, 2017

Hey,

thanks for your time doing those tests. I do agree with the other comments here who state that I would not expect to see much of an advantage in sampling 10bit after going through such a heavy compression to 150Mb in 4K … any detail that might be picked up is long lost after going through this.
Codec and compression in my eyes are way underestimated by many people.
This goes for color as much as for resolution.
In reality at 150Mb there’s not that much of your 4K resolution left and if you compare a 2K Amira ProRes4444 to internal 4K C300 MKII footage you will probably find more detail in the blown up 2K Arri footage than in the 4K Canon XF-AVC image.

One note about the metabones speedbooster – you don’t get a shallower depth of field from using a speedbooster, which is basically a focal reducer and makes your lens wider – it doesn’t change your depth of field for the same field of view.

Cheers,
Dan

 Ed Lapeos
Ed Lapeos
Member
March 2nd, 2017
Reply to  Dan Edmondson

Not really true that a speedbooster doesn’t change how shallow DOF one gets.

Because a lens used with speedbooster changes to a wider angle of view you have to move closer to get the same field of view as without the speedbooster and in doing that DOF becomes more shallow if you focus on the same subject.

 Dan Edmondson
Dan Edmondson
Member
March 2nd, 2017
Reply to  Ed Lapeos

That’s why I said for a given field of view. If you use a wider lens without the speedbooster at the same distance you get the same depth of field.
So if you chose the shot you want – meaning your point of view and your focal length – the speedbooster will not change your depth of field because for the same shot without the speedbooster you’d use a wider lens.
If you move the camera in to compensate for your wider field it’s not the same shot any more.
So for the same shot the speedbooster won’t give you a different depth of field because effectively it just adds another element to your optics resulting in a wider field of view. With the same field of view on another lens with the same effective focal length you get the same depth of field at the same T/stop. That’s it.
That’s the reason why people went to great lengths to build 35mm DoF adapters with screens in them because just chucking another element to your optics doesn’t affect DoF. DoF for a given distance only depends on field of view, sensor/film size and f-stop.

Eno Popescu
Eno Popescu
Member
March 2nd, 2017
Reply to  Ed Lapeos

From you theory, there should be no differences in DOF from a medium format camera to a cell phone one.

Using a speed booster man gets shallower dept of field, if he desires!

 Momir Alvirovic
Momir Alvirovic
Member
March 2nd, 2017
Reply to  Dan Edmondson

Having 50mm f1.8 lens on focal reducer with x0.71 reduction – makes it exactly the same as 35 mm f1.3 native lens. Having 35mm f1.4 lens on x0.71 reducer – makes it exactly the same as native 25mm f1.0 lens.
As majority of native m4/3 lenses do not have so wide aperture, it is possible to get shallower DOF and more exposure by using these reduced lenses.

You could look at same thing from other direction, which is my personal favorite (I use reducer all the time) – m4/3 camera with x0.71 focal reducer makes pictures of same FOV and DOF as Super35/APS-C camera – so putting 35mm 1.4 lens on GH5+x0.71 gives same FOV and DOF as putting it on Super35. In the process su still get 1 stop brighter picture at same ISO due to the reduction…

Member
March 2nd, 2017

One would hope this is a reflection of pre-release firmware.

I’d love to see this test done using an external recorder, to eliminate any issues caused by the CoDec’s compression levels.

Personally, I’ve never warmed up to V-log, IMHO & personal experience, Cine-D gives more usable range from the Panasonic products.

 Tom Barrance
Member
March 2nd, 2017

Really interesting, looks like another camera with great specs but maybe not such great real world usability. But it would be interesting to compare it with other cameras in the same price range. I’d be buying it for 1080p, not 4K. I’d like to see how it compares with more affordable Canons like the XC15 (which has a log profile) and the C100.

Joseph Moore
Member
March 2nd, 2017
Reply to  Tom Barrance

This article all but ignores real world usability, which is where the GH-line has ALWAYS excelled.

The title of the article should have been “h264 compression doesn’t work well with LOG gamma images.” That’s not a news flash.

Joseph Moore
Member
March 3rd, 2017

Sebastian,
Respectfully, you are wrong.
This isn’t a matter of opinion. Macroblocking (which by its very definition, averages similar colors) is introduced at the h264 compression stage. Period.

The codec is doing what it is supposed to, but as a delivery technology, it was not designed to be used with pre-compressed LOG images. When expanded in post, the damage has already been baked in.

 Matt Wolcott
Member
March 4th, 2017
Reply to  Joseph Moore

If this is the case, do you think the 400mb compression coming down the road will be enough to fix it? I have the camera on pre-order and now I’m second guessing myself.

Also I was wondering if you have a link to your site (I’m assuming you have a site where you talk about this kind of thing, from an earlier comment by Mike Hannon).

Member
March 4th, 2017
Reply to  Matt Wolcott

Buy from a reseller that has a 14 or 30 day return policy. Rent an external recorder and see how you like it. What if you love how it looks to you right out of the box?

All the anxiety here could have been avoided by waiting until the 400Mbs was released and the camera could have been reviewed by more people. Being the first to “scoop” what may or may not be an issue drives traffic to your web site, but in this particular case it may not have served some of the C5D readership very well.

All of this now has you 2nd guessing yourself when this camera has not even shipped yet and the conclusions drawn here are from only this one source. You may just love your new camera. In which case all of this may not mean anything to you in the end.

Joseph Moore
Member
March 6th, 2017
Reply to  Matt Wolcott

No. The all-i codecs uses the extra bandwidth to record every frame so that there is less CPU load when editing. You should not necessarily expect a PQ increase.

 Matt Wolcott
Member
March 6th, 2017
Reply to  Joseph Moore

How incredibly disappointing. I spent a lot on my computer so it can handle intensive calcs with 4k footage. I’d much rather have the extra bandwidth improve image quality. The IPB format looks fine to me and I have no issues editing it.

MAN.

 Mike Hannon
Member
March 3rd, 2017

You keep saying this and I don’t know why. There is a bucketload of evidence showing that high bitrate, externally recorded GH4 VLog does not suffer from these chroma smearing artifacts to the same extent as internal low bitrate H264. Where is your evidence that it’s unrelated to compression? If you have none then your statement is an opinion, a feeling of some sort, without empirical basis.

Eno Popescu
Eno Popescu
Member
March 2nd, 2017

Sorry but I really didn’t enjoyed reading your review, it’s more flawed than any so called flaws you’ve “uncovered” in the GH5 camera.

Where should I start?

DR: its actually more than 12 stops but the fact that you liked to disable the NR an then complain about some noise in the shadows makes me wander a lot of things.In fact after downloading your DR test, I brightened up the shadows and cleaned them a little bit and the DR was in fact on par with the XT 2, around 12+ stops!

Soft 180 fops? You didn’t complained that much of any Canon DSLM you’ve tested in your lab about the very soft 30fps 1080p. And by the way, the 180 fps on the Panasonic have more res compared to all the crappy Canon DSLR 1080p.:)

Noise: You’ve parsed how the XT 2 is so good at high ISO, and the Gh5 is comparable in fact but you say, naa, is not so great. Actually it is!

About the 10 bit testing: You’ve commented there is no real advantage in shooting in 10 bit on the GH5 vs the 8 bit but never show in fact a single blue sky with heavily grading. The problem is I’ve personally graded some LOG 10 bits from the GH5 (with a sky component) and frankly it look fantastic. I couldn’t have done that type of grading with any other 8 bit camera. So who do I believe, you or my own ayes…My trust goes with the later.

About your so called problems with the Vlog on the GH5, have you asked yourself if your grading was at fault? Probably the LUT? Again I’ve graded GH5 LOG files with a skin tone component in the image and I couldn’t find the problem you’re describing.
To bad I can not post any images in my comment cause your entire “so called review” cold have been put under the ? sign altogether.

Johnnie Behiri
Admin
March 3rd, 2017
Reply to  Eno Popescu

Eno,

After you are done conducting your own test, don’t forget to publish it in http://www.43rumors.com/ and http://www.personal-view.com

I’m sure that many will benefit from your review.

Thank you.

Johnnie

Member
March 2nd, 2017

Comparing a camera that cost $1,999 to a camera about 14 x the cost is an Apples and Orange comparison. Also Canon color science is highly optimized for flesh tones. Canon freely admits they work very hard on flesh tones. Nothing news making there at all.

Oh and, – The “vertical aliasing test” is a actually a test of *horizontal* resolving power and aliasing ???
…… and a few other things

Some flaws here.

Johnnie Behiri
Admin
March 2nd, 2017
Reply to  Clayton Moore

Clayton Moore

Sorry to see that you failed to understand that the inclusion of the Canon C700 in this lab test is only for demonstration purposes so people can see how 10bit should have been looked like, so of course it is not a direct competition between the 2 cameras.

Hope it is clear now!

Thank you.

Johnnie

Member
March 2nd, 2017
Reply to  Johnnie Behiri

Johnny I understand that of course, a reference, I guess its just a difference of opinions on how this was positioned. :-) We camera geeks can tend to get mirrored in the bits and bytes and minutia of things. But thats part of what sites like C5D are for. Some commenters are taking it personally and thats silly. This post was guaranteed to bring lots of traffic and there is absolutely nothing wrong with that, lots of traffic and many opinions both ways.

In fact Simon’s comment below is pretty powerful for this reason. If you just shoot a scene and work to apply a high quality grade to each, (not zooming in at 400% or whatever) you would find that (price not withstanding) we are breathlessly close to a place where it does not matter what camera you use. Lets see what this looks like in 8 months down the road after some new firmware is applied and a wider swath of seasoned pro’s have shot with it.

 Mike Hannon
Member
March 3rd, 2017

Yeah, but the “vertical tube” has horizontal lines in it… even I, an unscientific person, can see that. I mean, if someone points out an error you’ve made, why not say “Oh OK, I didn’t know that” instead of trying to cover your tracks and discrediting yourself further.

 Simon Bailey
Simon Bailey
Member
March 2nd, 2017
Reply to  Clayton Moore

Canon simply boost the red curve a little in the middle. Its not hard.

My a7s is better than the F5 I had. Camera prices mean nothing in relation to their quality to me anymore.

 Jeroen de Cloe
Member
March 2nd, 2017

Why would you want to use VLOG using internal recording anyway, at least at 100 mbit?
At this point compression becomes the bottleneck, especially with 10-bit 4:2:2. Just use an external recorder and store as ProRes! That’s where it’s meant for yet the article doesn’t test it.

And also, Panasonic is going to update the camera with 400mbit intraframe and that means internal recording will not show macroblocking anymore.

Thirdly, the grading in the example with the recording settings + in-camera stores is kinda ridiculous. Ofcourse you’ll see artifacts if you push an image like that. It’s not RAW.

Also, why are you reviewing a pre-production camera? The firmware isn’t even final.

Joseph Moore
Member
March 6th, 2017
Reply to  Jeroen de Cloe

Don’t expect the 400Mbps all-i codec to be a miracle worker. All of that extra bandwidth is used to make editing easier, not improve image quality. In fact, the GH4’s 200Mbps all-i codec looks *worse* than the 100Mbps ipb codec .

Member
March 2nd, 2017

A comprehensive test but ……..

My guess is, this test will be a good comparison up against the same information a firmware update or two down the road.
I remember the DVX200 V-log with it shipped was pretty bad. Just one firmware update took care of most or all or those initial issues. I wouldn’t take to much stock in what we see even before the unit ships.

Keep an eye on what Adam Wilt writes up on this over the next several months.

Michaël Marcopoulos
Guest
March 2nd, 2017

EOSHD.com is really not agree.
A test with a pre-production model.
Canon 1DC, 2,5 times more expensive.
FS7 is 4 times more expensive.
C700 is 14 times more expensive.

This article is really not fair, at all.

Member
March 2nd, 2017

I would say it is not a matter of being fair or unfair. It is about comparing the performance of this camera to other popular and high end cameras being used in the field. The GH5 has a price point and it touts several important and relevant features to video production. Consumers of the GH5 need to understand its limitations relative to the overall state of the art and to make value judgments before purchase. If it has serious flaws, then perhaps it is not a good purchase. However, the opposite could true wherein its performance is close to higher end alternatives, and therefore represents good value.

Member
March 10th, 2017

Not fair if this were a contest. It’s not, it’s a test to see how it compares to various standards. One day, you’re on set and you need to set up a “crash cam” below a speeding car. You’re not stupid enough t rig an Alexa or Red like that but based on tests some people had the good sense of doing, you decide to rig a GH5 to intercut with the high end rigs. Why? Because you know the image will hold up with the big boys fort he short time it’s on screen. And it’ll cut better than that go pro.

 Roxi Khalil
Member
March 2nd, 2017

I think it’s was high time this hype was taken down a notch. As a long term GH4 user I know too well it’s short comings. So with sheer amount of features squished (technical term) into the upcoming GH5 I was expecting a few short cuts to be taken. Your test do highlight similar issues to the origina GH4, dodgy skin tones and week slow motion. Now we mft users keep dreaming of an Sony killer to come and it does wound it to some degree. I was about to sell my GH4 and need a second camera for many reasons, The Spielberg’s out there wouldn’t have bought a GH5 any way, just the wishful thinkers, Panasonic are making great work horses for the majority of us, it’s for the skilled and steadfast to move up to higher echelons. Thanks for your continued unbiased views. Rox of ICE Entertainment signing out.

 Simon Bailey
Simon Bailey
Member
March 2nd, 2017
Reply to  Roxi Khalil

Actually Spielberg used some 5Dmk2s on that fighter pilot thing he shot. So yes the spielbergs might buy it.

Moses Wilson
Guest
March 2nd, 2017

I think people who never wanted a GH5 will still not want one and those who do or who have already ordered it will wait for an actual production camera with production firmware to arrive, figure out what settings produce the best quality image, and then conduct a test.

At this point everything is beta, the camera, the firmware.. Not to mention you’re looking at 10 bits after it’s been compressed down into H.264.

Sebastian Wöber
Guest
March 3rd, 2017
Reply to  Moses Wilson

Hi Moses,
I assume that firmware v1.00 is either final or near final. From 0.5 to 1.00 nothing has changed regarding our findings. If anything I hope our review will convince the manufacturer to fix the bugs in a future firmware update. The GH4 had similar bugs which were never fixed. Also, if I were to buy (pre-order) a camera I’d rather some reviewer points out those flaws before I’m sold.

Sean Badart
Guest
March 2nd, 2017

Yeah – keen to play with Steve’s when it arrives haha

Brayden Laffrey
Guest
March 2nd, 2017

Yeah I reckon I’ll stick to 8bit anyway

Member
March 2nd, 2017

I applaud this article. It is what we need to make a value judgement. I love the analysis of it. Supposedly Panasonic has a number of firmware updates planned for this year. So hopefully you all can do a re-evaluation this fall to see if progress has been made with regard to your issues. Thanks a bunch.

Miles Trotter
Guest
March 2nd, 2017

Silly review…You can’t judge a camera’s quality until it’s using actual production firmware and of course the 10 bit will suffer if you have a low bitrate(which will be fixed by a future update to higher bitrates)…

Sebastian Wöber
Guest
March 3rd, 2017
Reply to  Miles Trotter

Inclined not to answer a comment with such a disrespectful tone, but for the sake of clarification: Your assumptions differ from my experience after having tested over 50 different cinema cameras in our lab. Regarding firmware, I have never seen production firmware fix a camera’s quality in a way that would impact findings such as those we have found. If anything our findings will help and pressure the manufacturer to improve the product.

Miles Trotter
Guest
March 3rd, 2017
Reply to  Miles Trotter

I stand corrected. Thanks for your reply.

Gabriel Manz
Guest
March 2nd, 2017

-.-

 Momir Alvirovic
Momir Alvirovic
Member
March 2nd, 2017

Please note that physical size of sensor does not have anything to do with rolling shutter performance – as sensor scanning speed is not defined by millimeters per second, but by lines or pixels per second. If smaller sensor (while having the same resultion) would benefit, then A6300 would beat A7SII. Please be very careful with technical comments, as many people learn from your texts – it can be harmful for them to learn what is not correct.

 Luke Wen
Luke Wen
Member
March 2nd, 2017

Completely agree with you Momir, Cinema5D articles have made numerous factual errors over time. I remember in their A7S II vs A7R II lab test, they claim that “a7S II pixels can be larger and capture more light resulting in better lowlight performance.” This is completely nonsensical as the amount of incident photons are the same regardless of what cameras you use. Large photosites result in lower cumulative read noise which is the key to low light performance.

Cinema5D please hire a fact checker so you don’t mislead your readers.

 Simon Bailey
Simon Bailey
Member
March 2nd, 2017
Reply to  Luke Wen

Why do they need to hire a fact checker when they have you for free. But then who checks you? ;)

Joseph Moore
Member
March 6th, 2017

Sebastian,
Why don’t you go back and correct this article? It’s not like you published a book…you can fix at any time.

Member
March 10th, 2017
Reply to  Luke Wen

Luke, enlighten me. If a Full Frame sensor with the same megapixel count as an APS-C have pixels app 1.5x larger can you explain how it lowers the “cumulative read noise”? Also, I’m not so well verse on the tech terms here, but what is “cumulative read noise”?

Thanks

Manuel Flaggschiff
Guest
March 2nd, 2017

there is quiet a difference between 150 mbps and 400mbps in the future update so lets wait and see, as it’s just a beta firmware ;)

Member
March 2nd, 2017

Good point.

Sebastian Wöber
Guest
March 3rd, 2017

Hi Manuel, Actually the difference is not so big. The 400mbit codec will be an all-i codec, which requires much more data for the same amount of information. I don’t expect this codec to fix any of the problems pointed out, as those are most probably related to processing and not encoding.

 Ville Pakarinen
Member
March 3rd, 2017

“Actually the difference is not so big. The 400mbit codec will be an all-i codec, which requires much more data for the same amount of information.”

Maybe I’m not understanding, but aren’t these two sentences mutually exclusive?

 Momir Alvirovic
Momir Alvirovic
Member
March 2nd, 2017

Please note that blocking does not come from low bit depth but mostly as compression artifact. Having 10bit color and 422 subsampling at 150Mbps demands same compression as having 8bit 420 at 100Mbps – therefore amount of blocking artifacts happens to be the same in both cases, and therefore using 8bit 422 at 150Mbps introduces least compression so it holds up quite well…

 Momir Alvirovic
Momir Alvirovic
Member
March 2nd, 2017

Next, please note that V-Log L is by design very inappropriate for using with H264 codec (same as derived DJI D-Log), due its already compressed luma values (IRE 7.3 to IRE 61 for 0-90% reflectance, max IRE 80 for super whites), that is one more reason for blocking artifacts…

Johnnie Behiri
Admin
March 2nd, 2017

Hi Monir.

It can be that Panasonic should consider completely removing 10bit at 150Mbp and implement it only when 400Mbps comes around.

Thank you.

Joseph Moore
Member
March 2nd, 2017

This. Thank you!!!

 Lee Mackreath
Member
March 2nd, 2017

How much did metabones pay you to mention and hyperlink their hardware throughout this article?

Completely devalues anything you have to say about the camera… your tests are worthless

Johnnie Behiri
Admin
March 2nd, 2017
Reply to  Lee Mackreath

Lee.

No one is paying us anything to do those tests.

The hyperlinks are part of our system making it easy for anyone who wants to buy the mentioned product. Then, if one buys from the supplied link, we DO GET an affiliate commission.

Hope it’s clear. For more information, https://www.cined.com/about/

Thank you.

Johnnie

Member
March 2nd, 2017
Reply to  Lee Mackreath

Lee, Kind of a cheap shot.

Sam
Sam
Member
March 2nd, 2017

Nice write-up and well articulated, Seb! Though part of me can’t help but wonder if these tests were performed using an optically-adapted lens on the GH5. As good as Speedboosters are (I own the XL and love it), there is a greater likelyhood to loose a bit of resolution and even micro-contrast that is inherent in modern lenses – which ultimately could have an effect on skin tones and definitely would on resolving power. I would be curious to see how the tests would go if you repeated them with either native lenses (ie 12-35mm f2.8 or some PL primes) or if adopting lenses – at very least, use a non-optical adapter. Though please correct me if I am misunderstood – it wasn’t too clear if you used the same lens between the cameras (guessing you did), though by using any optical adapter (in this case a focal reducer), it’s no longer really the same lens when looked at as a whole – as it is 6 more objectives between the light and the sensor.

 Peter Gordebeke
Member
March 1st, 2017

How come this does not happen on other footage recorded in 10-bit V-Log?

Did some quit tests on footage others have made available and I could not reproduce this at all:

http://imgur.com/a/yQomh

I used the official Panasonic V-Log to Rec709 LUT you linked to, increased the exposure to 125%. When I looked at the footage of Emannuel Pampuri, with lots of close up skin, this was nowhere near as dramatic as you made it look!

Care to elaborate on that?

Also, why did you grade the skintone test to Rec709 yourself and not use the official LUT for that, as you did some lines later in the sky gradations test?

 Artyom Kornikov
Member
March 2nd, 2017

I clearly see distinct green and red patches on last three close ups on you images…

 Peter Gordebeke
Member
March 2nd, 2017

My phrasing was probably a bit too strong as well – yes, you’re right. You can see some patches, but it’s not to the same extent as displayed in the test from the review.

Joseph Moore
Member
March 2nd, 2017

This is h264 compression, people! It has nothing to do with 8-bit, 10-bit or 8 million-bit.

 Mike Hannon
Member
March 3rd, 2017

This comment is ironic. Cinema5D is presenting the less than complete understanding of the issue. This guy is trying to open your eyes.

Joseph Moore
Member
March 6th, 2017

You are completely, 100%, dead wrong. Plug in an Atomos or Pix-E and write me back.

 Simon Bailey
Simon Bailey
Member
March 1st, 2017

and what method did you use to export the comparison stills?

 Simon Bailey
Simon Bailey
Member
March 1st, 2017

Also what edit software did you use for this test? Are you entirely sure that you had the softwares settings setup for 10 bit.

 Simon Bailey
Simon Bailey
Member
March 1st, 2017

So I get to the end to read that the firmware is pre-production????

That makes this entire review worthless sorry. Very poor clickbait.

Joseph Moore
Member
March 8th, 2017

This is not a “bug”. This is expected behavior from h264. It’s literally how the codec is designed to work, and exactly the same reason why recording V-LOG L internally was so challenged on the GH4.

Is it unfortunate? Yes.

Would I have preferred a better codec at a bit rate more suitable for 10 bit 4:2:2 color? Absolutely.

Was that going to reasonably happen on a $2K camera from a Japanese consumer giant? Probably not.

Kamil Tamiola
Guest
March 1st, 2017

It looks your panasonic is simply doing numerical interpolation to 10bit. In other words there is a synthesised 2bit component. Shoot a sharpness or colour gamut card both in 8bit and 10bit modes, trim the clips to 10 sec exactly and compare file sizes :)

Nino Leitner
Guest
March 1st, 2017
Reply to  Kamil Tamiola

Dude you lost me ?… Tell SSebastian Wöber

Kamil Tamiola
Guest
March 1st, 2017
Reply to  Kamil Tamiola

Nino Leitner Robby, unless compression took the details away, 8bit and 10bit shots you boys have shared on C5D look pretty much the same. I say unless, as my dear kids on the block web JPEG container is 8bit by default, hence I wouldn’t be surprised if Panasonic will spank your sorry asses and ask you to generate TIFF dumps from your videos (uncompressed, 16bit). My educated guess is (still a guess), the camera can’t shoot ture 10 bit in the whole color range, hence the crappy 8bit like output from 10bit. How do you compensate for that? Simple, “predict” the missing color information back :) Panasonic wouldn’t be the first “more than 8bit” camera to do that! :D Is that more understandable ?

Joseph Moore
Member
March 2nd, 2017
Reply to  Kamil Tamiola

I know of no 10-bit camera for which this is an issue. The camera actually captures at higher than 10-bit. This IS a common issue with display panels, though.

Nino Leitner
Guest
March 1st, 2017
Reply to  Kamil Tamiola

Haha okay ? yep a little bit, my friend

Kamil Tamiola
Guest
March 1st, 2017
Reply to  Kamil Tamiola

Nino Leitner Splendid!

 Ville Pakarinen
Member
March 3rd, 2017
Reply to  Kamil Tamiola

What a turn off if that’s true. I guess what you’re describing only applies to internal recording, not external? Could it still be an amazing 10-bit camera with an external recorder?

Sebastian Wöber
Guest
March 3rd, 2017
Reply to  Kamil Tamiola

Hi Kamil,
Correct me if I’m wrong, but I think if the GH5 were to “guess” the missing 2 bits, we would see a gradient between the two adjacent color fields.

Kamil Tamiola
Guest
March 3rd, 2017
Reply to  Kamil Tamiola

Sebastian Wöber Only if your “guessing” algorithm assumes the use of gradient. Secondly, you may have an infinite number of gradient algorithms: line, parabolic… etc etc… Once again, this is a guess. Without an exact data in my hand, treat my comment as an informed opinion.

Richard Dimery
Guest
March 1st, 2017

Is 10 bit only available in 4k!?!

Johnnie Behiri
Admin
March 2nd, 2017
Reply to  Richard Dimery

Currently yes

Matthew Manteiga
Guest
March 1st, 2017

Brutal

 Chris Maldonado
Chris Maldonado
Member
March 1st, 2017

Sebastian, thank you so much for this article. I love how you guys at Cinema 5D do not let any marketing BS or any other tropes of the industry persuade your judgment when it comes to sharing the results of your tests with the world. I appreciate your sincere and thorough examination of the still beta preproduction firmware variant of the GH5 software and VLOG. I currently shoot with a PIXE5 and VLOG all the time and find it to be quite easy to grade and definitely usable professionally if you know how to properly make the footage bend to your will haha. That said it does take a bit more work than I’d like, so hearing that the GH5 could potentially have some of the same problems as the GH4’s already available 10bit 4:2:2 VLOG [externally] is slightly disconcerting. I do hope Panasonic fixes some of these issues but am thankful for the LUT you guys provided. I picked up your LUT way back when for the sony cameras external SLOG issue thru the HDMI and was happy then, so I’m sure this will help with the GH5 now. Thanks for your hard work, cheers! -Sully Cortez

 Mark Aylward
Member
March 1st, 2017

I wish you could have place a Black Magic camera up against this camera…

 Greg Althoff
Greg Althoff
Member
March 1st, 2017

It’s always a bad sign when a company says you have to wait months for the camera to work at it’s full potential.

Besides that I really don’t know why this camera is being hyped. It’s proving to be extremely underwhelming vs the competition from even 2 years ago let alone this year. To me this would be a waste of money, especially knowing that both BlackMagic and Sony are about to release new models that will most likely be the future of small compact video camera’s.

Member
March 1st, 2017
Reply to  Greg Althoff

I’m not sure that’s really fair. Both Sony and black magic have had extensive issues on launch (over heating, black sun) Black magic love to launch late and often with numerous faults requiring long waits for firmware updates. As for being two years behind the curve? What are you comparing this too?

I’m sure you’re right though, no doubt Sony will throw a curve ball.

Member
March 1st, 2017

This problem with 10-bit and V-Log was exactly the same on the GH4 shooting 10bit with an external recorder. V-Log is an absolute debacle. It took forever to arrive, it’s unuseable because it’s blocky/banding-crazy, and it’s ridiculous that you have to buy it in a cardboard box.

I sold my GH4 for a C100 MkII with BM Video Assist and yes, specs on paper do not a good image make. The C100 is still a better camera.

Member
March 9th, 2017

As I understood it with V-Log on the GH4 – and have no reason to assume it’s any different now – all Panasonic did was take their Varicam v-log and chop off the last few stops of DR. It’s a log profile that just wasn’t constructed for these 11 stop cameras.

If you want me to dig out my 10-bit GH4 v-log banding test clips I can have a look for them. Pure white walls were showing dancing pink and yellow circles!