I've had much success with both codecs. I haven't found either to be an issue.
I shot this review with 10 bit UHD All-Intra. Edited so much better on my older Mac Pro. With Long GOP I have to use proxy mode.

https://youtu.be/Kz0vq1uubKk

I've had much success with both codecs. I haven't found either to be an issue.]

No problems here either.

This EVA1 short (granted, 16:9 UHD instead of DCI 4K) was shot with the same exact LongGOP codec (150Mbps) in full Vlog (not the 12-stop constrained Vlog-L variety), *and* uses a hazer for the radio station scenes. If anything is going to break LongGOP, this is it.

Shot on EVA1 | Short film "Radio 88" on Vimeo

Im really not sure why Panasonic chose these 150Mbp/s specs for this 10bit 4:2:2 CODEC.

Campared to the 8bit 100mbp/s 4:2:0 cousin, the 150 has the burden of TWICE the chroma resolution and arround 40 times larger color palette....and all it was given was 50 additional megabits?

Ironically, the compression ratio is WORSE on the 150 than it is on the 100. ( strictly mathematically speaking)

I did some of my own blue sky tests using the 150 and VLog. I graded it to 709 and let me tell you, it didnt just "band" in any way, it downright "macro-blocked" badly in the gradient blue shades. For me, that 150 broke like a cheap wine glass. The same exact shot with ProRes over HDMI was bulletproof. I pushed that ProRes WAY harder and got ZERO artifacts.

Its being said that all three Panasonic demo videos used the 150 CODEC. And people wonder why there are so many issues with it? Mitch Gross of Panasonic of all people, fully understands the value of shooting ProRes over massively compressed long GOP. He spent years ar Convergent Design explaining this to everybody.

For me, after what I have seen that 150 do?...im never using it with VLog. Im OK with 100, 8bit for CineD or Natural 709-ish profiles and will only shoot VLog on ProRes or internal 400 All-I.

Its being said that all three Panasonic demo videos used the 150 CODEC. And people wonder why there are so many issues with it?

What issues are those exactly?

Check the comments under every video. You will see quite a huge number of negative observations from a wide range of people. Its important to note that of the few EVA1's in circulation right now, Panasonic has recently placed a mandatory halt on anybody releasing footage from their evaluation unit. It seems that maybe Panasonic has taken the high amount of negative feedback to heart and are making changes to answer the critics? I have a slight "wild" hunch that the EVA1 might be delayed to rework some of the noise reduction processing? I dont know, we'll see. If I were Panasonic, I REALLY would not release this camera without the 400mbp/s All-I CODEC on day 1.

Im really not sure why Panasonic chose these 150Mbp/s specs for this 10bit 4:2:2 CODEC.

Campared to the 8bit 100mbp/s 4:2:0 cousin, the 150 has the burden of TWICE the chroma resolution and arround 40 times larger color palette....and all it was given was 50 additional megabits?


The extra 50 mbps should pretty much cover it.

Each UHD 10 bit 4:2:2 frame before encoding requires only 1.667 times the space compared to an 8 bit 4:2:0 frame. The ratio between 150/100 is only 1.5 but a 10 bit resolution actually compresses better than an 8 bit resolution.

Also 4:2:2 should compress fine because remember that the original pre de-Bayered data from the sensor only requires half the space compared to the de-Bayered data.

Check the comments under every video.

I have. The Vimeo versions. The versions that didn't get screwed up with the encoding like what happened on YouTube. Do you subscribe to the idea that YouTube is the best representation of uploaded material? It was explicitly stated in the roadshow last night that the YouTube versions didn't looks as nice as the Vimeo. Have you checked the comments on the Vimeo versions? Which comment are you referring to exactly?

It seems that maybe Panasonic has taken the high amount of negative feedback to heart and are making changes to answer the critics?

That change would be introducing Intra earlier than next year. That's not the case.

I have a slight "wild" hunch that the EVA1 might be delayed to rework some of the noise reduction processing? I dont know, we'll see.

I'll be you $50 that the camera will be shipping at the end of this month.

If I were Panasonic, I REALLY would not release this camera without the 400mbp/s All-I CODEC on day 1.

But it will be and people will shoot some nice content in LongGOP.

So far, the only sample showing LongGOP issues didn't actually show issues with LongGOP. It showed corruption on the clip, after being ingested via FCPX. How about we see your clip that "breaks like glass"?

But it will be and people will shoot some nice content in LongGOP.

Right!

Or use 200Mbps H.265 using 4992 X 3744 in 10 bit!

Check the comments under every video.

I'll do you one better.

From "Radio 88" on YouTube.

This footage looks excellent. Easily the best looking picture of the three EVA1 films released.

Great Film! Will any of the short films shot on the EVA 1 be available for download? It would be nice to see what it looks like before YouTube's compression.

Lovely short film! I'd really like to know what lighting was used in the interiors shots and ISO, also I'd like to know if it was used a soft diffusion filter on the camera for interiors. Thanks! :-)

Beautifully shot and directed - but I can't help feeling like they were fighting the camera's flaws. Noise all over hell.

Here's a helpful tidbit: LongGOP doesn't make more noise. In fact, if LongGOP wasn't up to snuff, the noise would break it.

Way too much low light noise suppression. Motion echoes and noise is pretty bad here. Please dont blame youtube as there is more going on here then a Youtube issue. Image looks soft also.

"Motion echoes" is not a real thing. Noise is, once more for emphasis, not a codec breaking. Image doesn't look soft, I saw it both projected and on an HDR 4K display.

Incredible job! Very cinematic. The camera really is exceptional. I believe that they have more latitude in improving the 2500 ISO and will do so because at this price point, this camera will rattle the market.

Not the right comments? Then find one yourself and point it out. This is from Radio 88, the short I was most impressed with. Also, the hazer, as I have mentioned before, should stress the LongGOP codec. So I have linked to the short on Vimeo, and here it is on YouTube:

Short film "Radio 88" shot on EVA1 by Johnny Derango | Panasonic - YouTube

Can you point out directly where the LongGOP 150Mbps codec causes problems?

OK,..just to make sure that nobody thinks I'm bashing the EVA1...

#1. I difenetely expect to love the EVA1 and I'm completely on track to buy it Q1 2018.
#2. I've been Sony guy my whole life and my GH5 has given me new respect for Panasonic today.
#3. YouTube or Vimeo video's are a terrible way to do a CRITICAL review of image quality. Nobody has to convince me of that!
#4. I do see sensor noise that is not from YouTube's crappy compression.
#5 . I dont feel like copy/pasting all the good and bad comments on this thread. We all know that opinions are like @ss holes, we all got one and they all stink. ;-)

However, I will say that I generally love all three of the videos there were made with Radio 88 being the best.. I can say that the western one, during the camp fire scene, there easily looks to have nasty temporal noise reduction artifacts in between each of the heavy flickering. (Frame 1 is clean, frame 3 or 4 changes so fast that the noise processing cant adapt quickly enough)

So, I see issues like that. I ignore the expected YouTube macro blocking and pixel clustering in the shadows. We all know that is not the camera's fault. However, there is clear fixed pattern sensor readout noise that is captured and displayed on some scenes and not at all on others.

Again, Panasonic has put a full and immediate block on any new footage being released from anybody with a sample camera while they continue to modify the camera processing today. I think we can see that Panasonic is clearly concerned about what they see and are moving fast to improve on what the complaints are about.

I'm confident that they will be able to deliver a great EVA1 to the public. Part of their tweaking might be them debating just how good they want the EVA1 to actually be. They want it good....but not "too good". They need to leave an acceptable "gap" between the EVA1 and the Varicam LT. There needs to be a noticeable "divide" between these models to justify each price point. Today, Panasonic might just be trying to find what that difference needs to be for marketing purposes.

One concern that I have is, although I'm sure they can deBayer and process a good clean video signal to the CODEC and the HDMI port, I'm wondering how the 5.7 raw sensor data output will look without the benefit of the in-camera processing and noise reduction processing.

As far as that 150 CODEC goes. If you own a GH5, shoot a deep blue sky in VLog, do a slow pan and grade it. You will probably see it,...not "band" like 8bit...but "macroblock" in perfect squares and clustered cubes between the shades of blue.

I think the 150 CODEC will look OK in the normal rec709-ish profiles even with the higher compression ratios.

OK,..just to make sure that nobody thinks I'm bashing the EVA1...

I really don't care if you are. If you're bashing the EVA1 for legitimate reasons, that's totally cool. If you're complaining about issues you don't understand and can't elaborate on, that's when I'm going to ask you to back it up. And if you can't, then you should just let that stand, and perhaps think about that if you have a similar post in the future.

#4. I do see sensor noise that is not from YouTube's crappy compression.

I don't understand why this is a point. No one is upset if you say the EVA1's released videos so far are noisy. I think that was the whole point (shooting some scenes at 5000 ISO). Your comment that is in contention is that LongGOP "breaks like glass" and is worthless, something you have yet to demonstrate.

#5 . I dont feel like copy/pasting all the good and bad comments on this thread. We all know that opinions are like @ss holes, we all got one and they all stink. ;-)

You're the one who, instead of pointing out specifically what *you* saw (because you hadn't seen anything at that point is my assumption) you brought up YouTube comments. So you decided to do that even though you agree that YouTube/Vimeo are terrible places to judge how the original codec behaved. Why did you make that call then?

However, I will say that I generally love all three of the videos there were made with Radio 88 being the best.

And, also, is the most stressful on the codec because of the hazer. You do realize that, right?

I can say that the western one, during the camp fire scene, there easily looks to have nasty temporal noise reduction artifacts in between each of the heavy flickering. (Frame 1 is clean, frame 3 or 4 changes so fast that the noise processing cant adapt quickly enough)

I have seen these shorts both projected digitally in a theater and on an HDR screen. There was no noise reduction that I can tell, hence why they look noisy (seriously, you bring up the noise, and now what to say there is NR being applied? So is the NR not working at all EXCEPT for one small part of the frame?)

Something else you don't seem to understand is that noise is terrible for uploading to services like YouTube and Vimeo because their compression doesn't deal well with it. What you're seeing is that failure. That's why adding grain to something that is going up on one of those services is a huge mistake.

However, there is clear fixed pattern sensor readout noise that is captured and displayed on some scenes and not at all on others.

I am very aware of what FPN looks like, and I haven't noticed that yet. What exact moment are you referring to?

I think we can see that Panasonic is clearly concerned about what they see and are moving fast to improve on what the complaints are about.

What concerns are those exactly besides noise? That's all I have seen so far.

As far as that 150 CODEC goes. If you own a GH5, shoot a deep blue sky in VLog, do a slow pan and grade it. You will probably see it,...not "band" like 8bit...but "macroblock" in perfect squares and clustered cubes between the shades of blue.

And I'm seeing that in bright red hair shot in Clog2 at 400Mbps Intra when saturation is applied. So your point is?

EDIT: Also, how am I grading this? With a LUT? LUTs can cause weird things to happen, especially in highlights. Are you talking about grading by hand?

Ugggg...c'mon Gary, I dont want to get dragged into deeper detail on this ;-)


______________________________________________

OK. [sigh].....So look at the camp fire scene at 6:00 minutes

Short film "Near to Superstition" shot on EVA1 by Elle Schneider | Panasonic - YouTube

This screams of struggling temporal noise reduction. Dont know if this is in camera or post. This is pretty ugly and I dont blame YouTube VP9 compression for this one. No way.
______________________________________________

First Official EVA1 Footage | Panasonic - YouTube

0:13 seconds - True sensor noise in the shadows at 800 ISO
1:56 - True, ugly sensor noise in the shadows.

Dual 800/2500 native ISO is supposed to be about the same noise amount. Supposedly?
_______________________________________________

I thought Radio88 looked the best but again, we dont know how much post processing was used. It was stated that the editors DID use it on some scenes.

As far as comments from people, there are plenty of good an bad to read on all three YouTube videos.

Yes, YouTube VP9 is a very low bitrate. Yes, you will see macro blocking and clustering in the shadows of ALL videos. However, it "can" reveal sensor noise, no doubt....and it clearly does in my opinion.

WAIT!.....Now Gary, you can say that everything I write here is 100% false. If you think it is, than let me just take back everything I said so we dont have to keep this "back and forth" stuff going.

YouTube is terrible and the noise we see is NOT from the EVA1 but 100% YouTube's fault.

YouTube's VP9 compression is incapable of revealing camera sensor noise.

The EVA1 pre-production sample bodies are not noisy in any way.

The 150mbp/s CODEC is perfectly fine and is not highly compressed at all and does not macro-block when pushed. My GH5 150 testing was an anomaly and it really holds up perfectly with 10bit ProRes.

People's comments are positive about the EVA1 and anybody with negative comments is clearly ignorant.

So, there. I retract everything I said for you. In reality, non of this really matters anyway until we see the the true and FINAL production EVA1's hit the street. None of this back and forth debate will matter in the future.

I think the final tweaked product will look good.

OK Gary, does all this now wrap up the discussion nicely?

Awesome,...whew,...OK! ;-)

CT

As far as that 150 CODEC goes. If you own a GH5, shoot a deep blue sky in VLog, do a slow pan and grade it. You will probably see it,...not "band" like 8bit...but "macroblock" in perfect squares and clustered cubes between the shades of blue.

I just did that, don't see those blocks you are talking about.

https://www.dropbox.com/sh/xofsa0519xzb1i9/AAD3LGvDKoTe5wJOOaIIhzEva?dl=0

"sky.mp4" is an out of camera v-log video and "sky graded.mp4" is a Rec.709 graded version with a bitrate of about 32Mbps!

So let's talk about those blocks. :)

This screams of struggling temporal noise reduction. Dont know if this is in camera or post. This is pretty ugly and I dont blame YouTube VP9 compression for this one. No way.

Did you compare it with the Vimeo version? You didn't compare it with the Vimeo version. Why didn't you compare it with the Vimeo version?

I think you're also confusing the natural motion of the fire on the group of ladies with bad NR. It's not, and it's very clear in the Vimeo version that it's the way the light from the fire is casting out onto the talent.

YouTube's VP9 compression is incapable of revealing camera sensor noise.

It's already been explained that VP9 compression struggles with noise, because it's fine random detail. It can cause all sorts of problems. That's what I meant when I said that's why it's a bad idea to add grain for a film look for something going to YouTube or Vimeo.

The 150mbp/s CODEC is perfectly fine and is not highly compressed at all and does not macro-block when pushed. My GH5 150 testing was an anomaly and it really holds up perfectly with 10bit ProRes.

Cary has done a test. So you can respond to that.

People's comments are positive about the EVA1 and anybody with negative comments is clearly ignorant.

#3. YouTube or Vimeo video's are a terrible way to do a CRITICAL review of image quality. Nobody has to convince me of that!

Didn't you say that? Now you're saying that their critical reviews ARE not a terrible way? So which one is it?

I just did that, don't see those blocks you are talking about.

https://www.dropbox.com/sh/xofsa0519xzb1i9/AAD3LGvDKoTe5wJOOaIIhzEva?dl=0

"sky.mp4" is an out of camera v-log video and "sky graded.mp4" is a Rec.709 graded version with a bitrate of about 32Mbps!

So let's talk about those blocks. :)

OK,...lemme take a look at it when I get home. I suppose I will do another test of my own too. My original test was done on the original firmware. I'm guessing you on 2.0?

Should be interesting....

Thanks!

To sum up everything I'm trying to say:

In my opinion;

The EVA1 looks like it's going to be a nice camera.

YouTube is a terrible way to judge a camera....however,...it still does reveal some the noise problems of the camera. This is not a contradiction. YouTube is not necessarily completely "worthless". It's good enough to show some of that sensor noise but you need to understand what is the camera and what is YouTube.

This could all be solved by Panasonic releasing a high bit rate download for people to analyze properly.

150 10bit is highly compressed h.264 and has shown me reasons not to use it in VLog

The EVA1 is still being worked on by Panasonic.

That's it. This sums up everything I'm saying in a nutshell. If I have said anything more to make you think something else, than just disregard it. I'm not building and extensive forensic, legal case here for the EVA1 or YouTube to be tried in a court of video engineering law here! Relax...lol

My observations are similar to others and maybe yours too.

Yes, I will look at the Vimeo versions as well as look at Cary's videos and I'll revisit the 150 10bit compression test too. It's time I compare 150 Long GOP with the new 400 All-I anyway too.

I really dont know why all of this back and forth quoting thing keeps happening.

CT

YouTube is a terrible way to judge a camera....however,...it still does reveal some the noise problems of the camera.

Which is not an argument a single person in this thread has made. Why did you bring that up?

This is not a contradiction.

It is because you have moved the goalposts. Now it's about noise. Before it was about LongGOP not working well. You need to stay on topic.

YouTube is not necessarily completely "worthless".

No one made that argument either, you're the only one to say anything close to that. What you're doing is confusing VP9 compression artifacting and confusing it with AVC 10-bit 422 LongGOP breakdown. That's been the point for the entirety of this thread. Why do you not understand this?

150 10bit is highly compressed h.264 and has shown me reasons not to use it in VLog

What is probably operator error has driven you to an erroneous conclusion.

The EVA1 is still being worked on by Panasonic.

And I bet you $50 (you haven't taken a single one of my bets) that it will ship with the same LongGOP AVC codec at launch that you're talking about, and no Intra will be available until 2018.

I'm not building and extensive forensic, legal case here for the EVA1 or YouTube to be tried in a court of video engineering law here!

You're spreading FUD about an issue which is stemming from what is probably going to work out to be a workflow issue on your part.

My observations are similar to others and maybe yours too.

We'll find out soon enough. By the way, Steven, who claimed to have seen similar issues to yours? His video doesn't show that at all.

Who made Huff a Trustee?

My God man....are you a lawyer or something? You are litigating this stuff way too seriously! Ease up dude, everything is cool. I'll get into the 150 testing soon with Cary. Lets see what happens.

Untill then, just try to ignore everytjing I wrote. Dont worry about it. Carry on,...as you were. Nothing to see here. ;-)

That's just the way Gary is. I just sit back and enjoy the show.

It's time I compare 150 Long GOP with the new 400 All-I anyway too.

Depending on the footage Long GOP compression is 2 to 3 times as efficient in compressing video with the same quality as All-Intra. So overall I suspect you will not see any significant difference between 150Mbps Long GOP and 400Mbps All-Intra.

Of course All-Intra edits much more snappy on slower computers but that has nothing to do with the quality of the uncompressed video.

Who made Huff a Trustee?

Regular Crew, Major Player, Trustee etc. are just the titles that get assigned once a member reaches a certain post count. In other words, it's just a designation the forum software system makes automatically and over time a member gets "promoted" by the changing titles. It's a function of post count and nothing more.

I'll be the first one to admit that those designations, now more than 17 years old, are pretty much irrelevant and I've been considering changing that space below the member name to show that person's business / company info / network affiliation / favorite My Little Pony character / or something.

I really dont know why all of this back and forth quoting thing keeps happening.

The forum thread page layout can be somewhat confusing. The button at the bottom right will quote the previous post in full. The button at the bottom left will NOT quote the previous post.

Their labels, "quote" and "post reply," respectively, are not that well delineated graphically. The "post reply" button at the bottom left can be easily overlooked. I'm aware of that problem.

Meanwhile, for the broader audience:

If there's ever another member here whose posts you'd rather not see, just go to Controls > Edit Ignore List (direct link is http://www.dvinfo.net/forum/profile.php?do=ignorelist) and put in their name.

Some meta-discursive posts in this thread have been edited or completely withdrawn from public view. Please keep it civil and on-topic. Thanks,

As far as that 150 CODEC goes. If you own a GH5, shoot a deep blue sky in VLog, do a slow pan and grade it. You will probably see it,...not "band" like 8bit...but "macroblock" in perfect squares and clustered cubes between the shades of blue.

And Cary was kind enough to shoot his own test.

I just did that, don't see those blocks you are talking about. "sky.mp4" is an out of camera v-log video and "sky graded.mp4" is a Rec.709 graded version with a bitrate of about 32Mbps! So let's talk about those blocks. :)

https://www.dropbox.com/sh/xofsa0519xzb1i9/AAD3LGvDKoTe5wJOOaIIhzEva?dl=0

And now I have one of my own. DCI 4K 24p, Vlog-L, 10-bit 4:2:2 in both LongGOP and Intra. Brought into DaVinci Resolve Studio 14 and corrected (no LUT). I used the color wheels, adjusted Lift and Gain values (didn't touch the mids because it didn't really need it), and did hue transforms for the sky, both hue and luminance in the blues.

https://drive.google.com/open?id=0BxrUdQWZEmysYURxNkxvTTJXX28

The clips are randomly placed in the timeline between the Intra and LongGOP versions. So which are which? According to Cliff, the LongGOP version will "'macroblock' in perfect squares and clustered cubes between the shades of blue." So it should be easy to tell. Make sure to download the original QuickTime ProRes version.

Depending on the footage Long GOP compression is 2 to 3 times as efficient in compressing video with the same quality as All-Intra. So overall I suspect you will not see any significant difference between 150Mbps Long GOP and 400Mbps All-Intra.

Of course All-Intra edits much more snappy on slower computers but that has nothing to do with the quality of the uncompressed video.

Cary, I grabbed your file and tossed a VLOG to Rec709 800% LUT on it I made form LutCalc.

I think it looks OK. I'm not seeing any blocking. Here is your file that I tested. It was done in a 32bit color space.

part 1 - was your clip ungraded
part 2 - is my 709-800% LUT added
part 3 - is a color canceled luma grayscvale filter. This is "looking under the hood" of the CODEC. You can see the Long GOP IPB cadence at work. Oddly enough, there are some straight line and block artifacts on the left side of the image but it's fully covered up under normal RGB viewing.

It's looks OK!

Now, I tested my camera on the original GH5 firmware a while back. I did it twice on two different days and got some ugly blocking artifacts. I have not tried it since then and we have gotten two firmware revision since. I'm guessing you are 2.0 today?

I'm bringing my GH5 to work tomorrow and shooting some new 150 CODEC footage and run it through the ringer. Thanks for the upload.

https://drive.google.com/open?id=0B60OfAOTK8ZtV0JaUnFHMzFvWlk

CT

Now, I tested my camera on the original GH5 firmware a while back.
1.0 definitely had some issues (ghosting) with v-log, issues that were fixed in 1.1

I don't know if any things were fixed (under the covers) for 2.0, but I do like the colors coming from the 4992 x 3744 10 bit 4:2:0 V-Log encoded with H.265.

part 3 - is a color canceled luma grayscvale filter. This is "looking under the hood" of the CODEC. You can see the Long GOP IPB cadence at work. Oddly enough, there are some straight line and block artifacts on the left side of the image but it's fully covered up under normal RGB viewing.

This is not actually a thing. You can use a channel splitter which takes YUV and splits it into what the RGB components are, but you will not see the LongGOP IPB "cadence" (again, not a thing). In fact, I'd say what you're calling the "banding" (it's not banding), is actually some kind of render error from whatever filter your adding to that. Something in AVISynth?

EDIT: Yes, when I split out the signal into individual RGB components, the lines from that rendered aren't there for me in Resolve Studio 14. I'm going to say that since you're trying to concoct a LUT via LUTCalc, that your issues are stemming from that.

I don't know if any things were fixed (under the covers) for 2.0, but I do like the colors coming from the 4992 x 3744 10 bit 4:2:0 V-Log encoded with H.265.

The ghosting was at high ISOs, which was probably some kind of software error involving high ISO NR.

None of the release notes mentions touching any modes that already existed, simply unlocking new ones, so I highly doubt there are any changes to the original recording modes.

None of the release notes mentions touching any modes that already existed, simply unlocking new ones, so I highly doubt there are any changes to the original recording modes.
4992 x 3744 10 bit 4:2:0 V-Log encoded with H.265 is new.

This is not actually a thing. You can use a channel splitter which takes YUV and splits it into what the RGB components are, but you will not see the LongGOP IPB "cadence" (again, not a thing). In fact, I'd say what you're calling the "banding" (it's not banding), is actually some kind of render error from whatever filter your adding to that. Something in AVISynth?

EDIT: Yes, when I split out the signal into individual RGB components, the lines from that rendered aren't there for me in Resolve Studio 14. I'm going to say that since you're trying to concoct a LUT via LUTCalc, that your issues are stemming from that.

So that gray scale effect happens from a mixture of the 6 primary and secondary colors. (red, yellow, green, cyan, blue and magenta) The unsaturated parts are also included. This phase cancels out the color and brakes the CODEC in a neat way.

When I say you can see the IPB, what I mean is that you can actually see MPEG working long GOP calculations and plotting all around the image. This kind of phase cancellation reveals elements you cant normally see easily. This stuff is always somewhat hidden under normal circumstances. It's kind of useless but when you do this to ProRes HQ, you dont get artifacts like this. This is where complete I frames are superior. They dont need any data from previous frames to assemble a solid image. It's interesting to note that ProRes is a DCT-type system and is actually a relative of JPEG!

LutCalc wouldn't do those line artifacts to the image.

CT

What is your exact workflow for turning an image to grayscale? How does this process, by simply turning the image grayscale, invoke more information than splitting the video into individual R, G, and B elements and viewing each independently?

Doing a splitter node in Resolve does not yield any difference between LongGOP and Intra as you watch each channel isolated.

When I say you can see the IPB, what I mean is that you can actually see MPEG working long GOP calculations and plotting all around the image.

Please give a single, specific example of where one can actually see these GOP calculations happening in the video you uploaded.

4992 x 3744 10 bit 4:2:0 V-Log encoded with H.265 is new.

Yes, it’s a new mode that has been added, not a previously existing mode that has been updated.

Not really sure I’d ever want to shoot in that, being 4:2:0.

Please give a single, specific example of where one can actually see these GOP calculations happening in the video you uploaded.

So you see pixel clusters and shapes moving around the flat, non detailed portions the black and white image I uploaded? This is part of the Long GOP process. It's recycling "crumbs" from frames before each other.

And no, before you complain..."crumbs" is not a legal, technical, or "official" MPEG Long GOP term. It's my very own created word to describe the data carry over.

Haven't you ever taken a ProRes HQ master file, then compressed it into H.264 or MPEG 2 and re-applied that new copy (frame sync'd) on top of the original master with a phase inverted filter to reveal (in a gray scale) the Long GOP artifacts that the original master does not have? I'm pretty sure everybody has done this before for fun.

CT

Haven't you ever taken a ProRes HQ master file, then compressed it into H.264 or MPEG 2 and re-applied that new copy (frame sync'd) on top of the original master with a phase inverted filter to reveal (in a gray scale) the Long GOP artifacts that the original master does not have?

What is your workflow exactly for applying this phase inverter filter? What is the software you are using and what is the filter?

Any NLE can invert a waveform. You are flipping the wave "inside out" Vegas 14, Premiere, anything.

Take two virtually identical sine waves. Make even the slightest frequency change or phase shift to one, then invert it and apply back it to the other. The parts that are in phase will disappear because of cancellation. They only parts that exist will be the EXACT elements that were altered in the second one.

It's called "phase cancellation". Even phase canceling color channels and blending the inverted waveform will wipe out wave shapes and reveal others.

I know that a man as smart as you Gary, already knows this. So why are you asking me to explain it to you?

Time for bed!

CT Zzzzzzzz

Yes, it’s a new mode that has been added, not a previously existing mode that has been updated.

Not really sure I’d ever want to shoot in that, being 4:2:0.
What is the problem with that?

This "open gate" 5K mode takes almost the full sensor resolution. The full sensor has just as much (but differently formatted) information as 4:2:0, there is simply not more information available.

When you scale 5K 10 bit 4:2:0 video to 75% to make it UHD you get an approximate equivalent amount of information to 4:2:2. Theoretically you could even improve the bit resolution!

Plus you have more flexibility in framing in case you need to because you record in 4:3! And 2.0 offers aspect ratio markers in display!
Also this mode effectively gives you an equivalent 400Mbps Long GOP mode!

This is definitely my favorite mode to record in, and the colors look fantastic!

If you have not tried it I highly recommend it!

Actually, you can’t. Better to let the GH5 do the processing. There is no contest between recording 4K 4:2:0 and downscaling into 1080 (for the theoretical 4:4:4) and capturing in 444 RGB mode. The real world results simply aren’t there.

Also, bit precision absolutely will not increase with a downscale at all. That’s done at the capture level, even before debayering.

There is no contest between recording 4K 4:2:0 and downscaling into 1080 (for the theoretical 4:4:4) and capturing in 444 RGB mode.
There is nothing theoretical about it.

4:2:0 4K color (CbCr) information is already at 2K, so by just down scaling the luminance (Y') 50% you automatically get 4:4:4. What is incorrect is that the bit depth increases 4 times as well (due to oversampling), actually only the luminance component increases the bit depth with a factor 4, the color info actually remains at 8 bit because it is actually untouched with the down scaling.

Feel free not to use the 5k mode, I think you miss a lot of opportunities. :)

Yes, the open gate mode is awesome and even at 4:2:0, you still have a huge amount of chroma resolution. However, the bit depth cant change. I see what you are saying about pixel scaling but whenever you take two or more 10bit values and mix them or sum or bin them, you will always get a new 10bit value.

Now if you took one pixel with a 10bit value and combined it with another pixel that had a 12bit value then you could achieve a new 12bit sum. But you could mix and match all the 10bit values you want together but you will always end up with a new 10bit product.

Here is another way to visualize this. Imagine millions of pixels but each one having only a 3 color scale. (3 color shade possibilities)

0 = pure black
1 = middle grey
2 = pure white

You could mix millions of these 0, 1 and 2 pixels as much as you want in a 32bit color space but you will never get any shade more than black, grey and white. You would need to introduce a value that is "outside" of that 3 shade scale to offset these 3 numbers to produce values that are beyond 0,1 and 2. (like tossing in a "1.3" grey pixel or a "0.5" grey or a "1.9" white pixel...but those pixels you just dont have)

But yeah, I have seen some fantastically detailed videos shot with the full 4:3 sensor. Check this one out its CRAZY detail. You can literally make out faces of people way off in the distance.

�8K】Night Shibuya in Lumix GH5 6K mode and aspect 4:3 - YouTube

There is nothing theoretical about it.

There is if you cannot demonstrate any benefit in color correction over 444 captured directly. My own internal tests do not demonstrate any gain with common workflows.

Feel free not to use the 5k mode, I think you miss a lot of opportunities.

Opportunities like what exactly? First, the 5K mode is 4:3, unless open gate is making a comeback, that's already a non-starter.

�8K】Night Shibuya in Lumix GH5 6K mode and aspect 4:3 - YouTube (https://www.youtube.com/watch?v=PPnFaCZSjcc)

The YouTube version you linked to isn't even 4K. It's only in 1440p. The GH5 already uses the entire sensor to downscale and does a good job at it, far more than YouTube's VP9 compression, so unless you have a direct download source for this video, the link itself absolutely does not show an incredible level of detail.

#3. YouTube or Vimeo video's are a terrible way to do a CRITICAL review of image quality. Nobody has to convince me of that!

There is if you cannot demonstrate any benefit in color correction over 444 captured directly. My own internal tests do not demonstrate any gain with common workflows.

Opportunities like what exactly? First, the 5K mode is 4:3, unless open gate is making a comeback, that's already a non-starter.
Gary, I think there is not much of a point in having a discussion with you, it seems your teacup is already overfull!

Gary, I think there is not much of a point in having a discussion with you, it seems your teacup is already overfull!

Well, I wasn't quite looking for a discussion. I made a point that you won't see a difference if you break 4:2:0 in 4K, it will still break the exact same way even if you apply the same correction after a downscale to 1080 in a theoretical 4:4:4 mode. I have done this testing. If you have as well, and got a different result, I'd love to hear about it.

Here is another way to visualize this. Imagine millions of pixels but each one having only a 3 color scale. (3 color shade possibilities)

0 = pure black
1 = middle grey
2 = pure white

You could mix millions of these 0, 1 and 2 pixels as much as you want in a 32bit color space but you will never get any shade more than black, grey and white.
Not true.

For instance if you have four sensors they could record 0, 0, 0, 1 the resulting oversampled value would be 0+0+0+1 = 0.25. You would get a resulting 0, 0.25, 0.5, 0.75 and 1 as all possible values.

A sensor has a margin of error in recording a signal value which is called the S/N ratio, if you multisample values multiple times (at the same time) you increase the S/N ratio and thus the bitdepth.

By the way here is a 'grey' image using only black and white pixels:

https://i1.wp.com/nikles.it/wp-content/uploads/2017/07/Untitled-1.png

I get the math, I get you can do it with a grayscale image. However, "real world results" means, what benefit can you gain from an actual recorded piece of video in color correction, UHD 4:2:0 vs 1080 4:4:4. And the benefit in grading doesn't translate over. I have done this test.

Well, I wasn't quite looking for a discussion
I know, you only lecture.

I know, you only lecture.

You have more than enough opportunity to demonstrate that I'm wrong, just like Cliff does. However, you want to complain about my tone and that I don't just take your word for it.

I'm not interested in just taking your word for it. I want to back up your claims with evidence. You won't even respond about if you've tested this yourself. How am I supposed to respect your opinion given that?

Not true.

For instance if you have four sensors they could record 0, 0, 0, 1 the resulting oversampled value would be 0+0+0+1 = 0.25. You would get a resulting 0, 0.25, 0.5, 0.75 and 1 as all possible values.

A sensor has a margin of error in recording a signal value which is called the S/N ratio, if you multisample values multiple times (at the same time) you increase the S/N ratio and thus the bitdepth.

By the way here is a 'grey' image using only black and white pixels:

https://i1.wp.com/nikles.it/wp-content/uploads/2017/07/Untitled-1.png

That's interesting Cary. It's confusing,..but interesting! But how would that result in real world color? I'm guessing that would produce arbitrary weird colors???

That's very interesting! I wonder if anybody has actually scaled like this and calculated bit depth like that successfully. I'll check you link in a bit. ;-)

CT

You have more than enough opportunity to demonstrate that I'm wrong, just like Cliff does. However, you want to complain about my tone and that I don't just take your word for it.

I'm not interested in just taking your word for it. I want to back up your claims with evidence. You won't even respond about if you've tested this yourself. How am I supposed to respect your opinion given that?

I could make the statement; "It's a beautiful sunny day here in Miami Florida"

Gary would automatically say; "Nonsense,...display for me your exact GPS coordinates, barometric pressure readings, verified accurate temp readings, visibility distance measurements, humidity metrics and wind speed data from a source I trust."

Until then, he will argue with you until the end of time that you are speaking total B.S.!

CT lol....;-)