Page 47 of 49 FirstFirst ... 3743444546474849 LastLast
Results 461 to 470 of 481
  1. Collapse Details
    Senior Member Thomas Smet's Avatar
    Join Date
    Jul 2005
    Location
    Colorado
    Posts
    2,199
    Default
    Saying all DCT formats should look the same is like saying McDonalds, Wendy's, Culvers and Red Robin all have the same burger because they all grill ground beef and slap it in a bun.

    DCT is just a method to convert images to blocks of data. Two formats can absolutely be DCT based and have the same data rate and have a very different look. It has happened in the past (Photojpeg at 50mbps vs DV50 at 50mbps.). Even two different h264 or mpeg2 encoders can result in a different look and quality with the same data rate.

    The GH5 formats are odd little beasts and if the share anything with AVCIntra then they do use a form of motion prediction even for the all-I 400mbps format. Thats partially how 400mbps can get closer to the look of ProRes HQ at 800mbps. Half the data will eventually lose something in translation. That loss seems to come in the form of subtle gradients if pushed very hard.

    Wit that said the GH5 still is one of the few cameras that attempts to give users an affordable 10bit 4:2:2 format. No other camera under $5000 really does that. Perhaps the GH5 is not a grading beast in some situations but a lot of others have graded GH5 material with little to no problems. There is also the fact that not every videographer or even film maker grades at that level. The GH5 and GH5S still beat the pants off of other 8bit DSLR and video cameras out there when it comes to grading. Plus for the grading fanatics the GH5 camera do have HDMI out which is again something nobody else provides at 10bit besides BMD. Thats why I now consider Panasonic and BMD the only realistic players in the game.


    Reply With Quote
     

  2. Collapse Details
    Senior Member
    Join Date
    Dec 2012
    Posts
    240
    Default
    Does the Anamorphic mode have the same sharpening artifacts?


    Reply With Quote
     

  3. Collapse Details
    Senior Member
    Join Date
    Jan 2006
    Location
    Nashville, TN
    Posts
    1,132
    Default
    Quote Originally Posted by Thomas Smet View Post
    Saying all DCT formats should look the same is like saying McDonalds, Wendy's, Culvers and Red Robin all have the same burger because they all grill ground beef and slap it in a bun.

    DCT is just a method to convert images to blocks of data. Two formats can absolutely be DCT based and have the same data rate and have a very different look. It has happened in the past (Photojpeg at 50mbps vs DV50 at 50mbps.). Even two different h264 or mpeg2 encoders can result in a different look and quality with the same data rate.

    The GH5 formats are odd little beasts and if the share anything with AVCIntra then they do use a form of motion prediction even for the all-I 400mbps format. Thats partially how 400mbps can get closer to the look of ProRes HQ at 800mbps. Half the data will eventually lose something in translation. That loss seems to come in the form of subtle gradients if pushed very hard.

    Wit that said the GH5 still is one of the few cameras that attempts to give users an affordable 10bit 4:2:2 format. No other camera under $5000 really does that. Perhaps the GH5 is not a grading beast in some situations but a lot of others have graded GH5 material with little to no problems. There is also the fact that not every videographer or even film maker grades at that level. The GH5 and GH5S still beat the pants off of other 8bit DSLR and video cameras out there when it comes to grading. Plus for the grading fanatics the GH5 camera do have HDMI out which is again something nobody else provides at 10bit besides BMD. Thats why I now consider Panasonic and BMD the only realistic players in the game.
    Fair points. I do want to note that for the sample I posted, I would not consider it a particularly aggressive grade. Yes, it was a challenging scene in terms of dynamic range, but all I did was add a REC709 LUT (which you have to do for any LOG image) and adjust the curve from 50% down. For an actual deliverable, I would typically do a creative grade on top of that.

    Over the years, I've gotten in the habit of trying to avoid subtle gradation. It's become so ingrained in my thought process when designing shots that I almost forget that I didn't have to do that when I used to work on film. Yes, I'm really old. ;-) We shouldn't have to be afraid of clear skies and beige walls (oh my how h264 beige walls!) All tools have limitations, but if I can afford to shoot RAW in camera and avoid this one, it is really going to be liberating. It is great that the GH-series supports external recorders so well, but they do add additional cost, hassle and fail points. I am usually director and DP. Attention I have to spend on gear is attention I'm not spending on the talent.
    Last edited by joe12south; 05-16-2018 at 11:52 PM.


    Reply With Quote
     

  4. Collapse Details
    Junior Member
    Join Date
    Sep 2015
    Posts
    21
    Default
    Quote Originally Posted by Thomas Smet View Post
    I use an external BMD capture device via thunderbolt to capture HD uncompressed and it eliminates all of this. This is where I don't consider ProRes all that much more superior to the internal formats. As you have posted in some fine gradient situations ProRes can hold up better but from a VFX perspective both are about the same.
    Are you sure you get the results you get because of recording to the capture device (the Intensity as I presume). Using Natural sounds more plausible to me.
    I have the Intensity Pro and I see no difference between footage recorded to my Pix-EH5 in prores HQ and uncompressed 10 bot YUV to the Intensity card, besides obviously file size. From that I conclude that whatever is send over it doesn't outperform the quality of what ProresHQ can handle.

    Prores HQ/uncompressed 10 bit YUV/uncompressed 10 bit RGB (pick your flavor) is only marginale better (8 bit vs 10 bit) than the H.624 from the GH4. It still carries the same imprint from a blocking scheme, you so well described, after I choke it. If we really got clean data from the sensor those imprints should have been gone and that is not the case in any of the tests I ever did.

    Joe12South screen-grabs shows the same imprint in both the Prores and the H.624 PNGís.
    http://www.bunkweb.nl/plaatjes/prores_vs_h.624.png
    It would very much surprise me if the two would behave differently in their marching pattern when we look at a couple of frames in a row. Pretty much as we can see in the GH5S clip, of my YouTube attempt, on the back of the guy smoking a cigaret. So whatever the GHxx sends over itís not clean data. Maybe Joe is willing to test it or share a couple of more frames from that clip so we can test it ourselves.


    Reply With Quote
     

  5. Collapse Details
    Senior Member
    Join Date
    Dec 2007
    Location
    Viersen/Germany
    Posts
    172
    Default
    Quote Originally Posted by sgttom View Post
    Does the Anamorphic mode have the same sharpening artifacts?
    It doesn't.
    And in my tests less color noise and blotches as well.


    Reply With Quote
     

  6. Collapse Details
    Senior Member Thomas Smet's Avatar
    Join Date
    Jul 2005
    Location
    Colorado
    Posts
    2,199
    Default
    Quote Originally Posted by bunk View Post
    Are you sure you get the results you get because of recording to the capture device (the Intensity as I presume). Using Natural sounds more plausible to me.
    I have the Intensity Pro and I see no difference between footage recorded to my Pix-EH5 in prores HQ and uncompressed 10 bot YUV to the Intensity card, besides obviously file size. From that I conclude that whatever is send over it doesn't outperform the quality of what ProresHQ can handle.

    Prores HQ/uncompressed 10 bit YUV/uncompressed 10 bit RGB (pick your flavor) is only marginale better (8 bit vs 10 bit) than the H.624 from the GH4. It still carries the same imprint from a blocking scheme, you so well described, after I choke it. If we really got clean data from the sensor those imprints should have been gone and that is not the case in any of the tests I ever did.

    Joe12South screen-grabs shows the same imprint in both the Prores and the H.624 PNG’s.
    http://www.bunkweb.nl/plaatjes/prores_vs_h.624.png
    It would very much surprise me if the two would behave differently in their marching pattern when we look at a couple of frames in a row. Pretty much as we can see in the GH5S clip, of my YouTube attempt, on the back of the guy smoking a cigaret. So whatever the GHxx sends over it’s not clean data. Maybe Joe is willing to test it or share a couple of more frames from that clip so we can test it ourselves.
    Yes there is a difference. In particular when it comes to mosquito noise around edges which even ProRes HQ can display when you have edges of flat colors. Doesn't matter if you use Natural or log. DCT mosquito noise is DCT mosquito noise no matter what. ProRes HQ can macro block as well but its extremely rare. I see it a ton on ProRes LT and occasionally on regular ProRes. HQ is pretty robust however but it stills suffers from other DCT artifacts like mosquito noise. Remember ProRes is basically jpeg and while jpeg can look pretty good for general viewing it does still have artifacts compared to Tiff or raw photos. I'm also using the GH4 with its 1:1 sensor output and no extra processing involved. Perhaps the GH5 itself is causing that extra ringing or mosquito noise around the edges. I'm capturing external HD by letting the camera downscale UHD to HD via the HDMI port which results in the least amount of in camera processing.

    With that said all of this is really splitting hairs and few high end productions have ever considered ProRes HQ not good enough. In fact the ProRes4444 formats use the same level of DCT compression and have the exact same mosquito noise and that format is now used for a lot of Hollywood productions instead of raw. So we really are splitting hairs now. If a movie in Hollywood that actually earns money at the box office is ok grading the compression of ProRes then I'm pretty sure it will work for all of us as well. The only difference between ProRes HQ and ProRes4444 is the added 4:4:4 color and 12bit color. Compression wise the macro blocks and mosquito noise is the same. You have to move to ProRes4444 XQ to reduce the compression amount which tends to be larger than what a raw format would provide. If we eventually get more ProRes raw options the ProRes4444 XQ would largely be a pointless format unless somebody actually uses a 3 chip 4k camera.


    Reply With Quote
     

  7. Collapse Details
    Senior Member Thomas Smet's Avatar
    Join Date
    Jul 2005
    Location
    Colorado
    Posts
    2,199
    Default
    Quote Originally Posted by joe12south View Post
    Fair points. I do want to note that for the sample I posted, I would not consider it a particularly aggressive grade. Yes, it was a challenging scene in terms of dynamic range, but all I did was add a REC709 LUT (which you have to do for any LOG image) and adjust the curve from 50% down. For an actual deliverable, I would typically do a creative grade on top of that.

    Over the years, I've gotten in the habit of trying to avoid subtle gradation. It's become so ingrained in my thought process when designing shots that I almost forget that I didn't have to do that when I used to work on film. Yes, I'm really old. ;-) We shouldn't have to be afraid of clear skies and beige walls (oh my how h264 beige walls!) All tools have limitations, but if I can afford to shoot RAW in camera and avoid this one, it is really going to be liberating. It is great that the GH-series supports external recorders so well, but they do add additional cost, hassle and fail points. I am usually director and DP. Attention I have to spend on gear is attention I'm not spending on the talent.
    I would still love to see if the GH5S fairs better due to a lack of in camera scaling. Sharper edges always encode with worse ringing and mosquito noise and have to work that much harder in general. Cinegal found the 400mbps on the GH5S to hold up very well compared to the ProRes on the Alexa Mini. Could the encoder be improved on the GH5S or could the different processing affect what the encoder can do vs the GH5? A lot of people have been fixated on low light from the GH5S and pretty much ignored what it meant to go back to a 1:1 sensor camera and what that could mean for more natural detail and how well the encoder can perform. I believe there are more subtle quality differences between the GH5 and GH5S we need to pay more attention to.

    In terms of the extra noise reduction on the GH5S at high ISO levels I think that is a given and using noise reduction in post on raw from the Pocket will likely kill just as much texture and fine detail. Noise is noise and once you get to a certain level you can make the image look clean but you will make it plastic looking in the process. Serious production should still use lights to attempt to keep the ISO at more realistic levels on the GH5S and on the Pocket. The one advantage of the GH5S is from a videography point of view what comes out of the camera is already cleaner without any extra needed processing. On a 4 camera shoot that means a ton vs doing noise reduction in post on a longer production which could take days to perform. Even if the post noise reduction can look a little bit better thats just something a lot of videographers cannot spend the time doing. The GH5S may have more aggressive noise reductio nat higher ISO levels but I think thats a good thing for the people that need the low light shooting the most.

    If a narrative production is shooting ISO 6400 just because they can then they have a lot more to worry about than just the plastic look affecting the professionalism of their work.


    Reply With Quote
     

  8. Collapse Details
    Senior Member
    Join Date
    May 2015
    Location
    Central NY
    Posts
    1,753
    Default
    Quote Originally Posted by joe12south View Post
    So, I'm not positive about this because it has been a very long time since I've actually read the ProRes whitepapers, but I think one of the distinctions between it and many other delivery codecs is that it is more even handed. h264, h265, JPEG, etc. rely on weakness in human vision to compress certain parts of the signal more than others (low contrast areas, and dark areas being the most obvious.)
    I believe that modern h.264 and h.265 are using PERCEPTUAL codecs in that they throw away the parts that we don't really (in theory) perceive, same process is applied to AAC and all of it's derivatives (HE-AAC, etc.). Then I think (again my memory is a little weak on this as it's been more than 10 years since I was researching this topic) that they go into the motion analysis to determine the other parts that need to be encoded (did the pixel move or change? if not then code it as a repeat from the previous frame). I never had a decent grasp on the DCT vs. Wavelet technology, but most of the devices I was using were DCT based (cameras with DV or DVCPro, etc.). Mr. Powell could probably school us on this.


    Reply With Quote
     

  9. Collapse Details
    Default
    Quote Originally Posted by Thomas Smet View Post
    With that said all of this is really splitting hairs
    Yes, you have zoom far in to see the difference. At least I do. Some have mentioned secondary color correction being hard, but I'm curious to see proof.

    Don't get me wrong, I would love to shoot only in RAW. It seems everything the manufacturer does to the image I would rather them not. But I'm waiting for memory cards to get to be around $30-50 for a terabyte (1 hour of 4K RAW).


    Quote Originally Posted by Greg_E View Post
    a decent grasp on the DCT vs. Wavelet
    Best set of videos I've seen explaining DCT, http://www.dvxuser.com/V6/showthread...n-Really-Works

    Approachable explanation of Wavelets from the guy who wrote Cineform, https://gopro.github.io/cineform-sdk/ (see Wavelet Transforms).


    Reply With Quote
     

  10. Collapse Details
    Senior Member Thomas Smet's Avatar
    Join Date
    Jul 2005
    Location
    Colorado
    Posts
    2,199
    Default
    Quote Originally Posted by combatentropy View Post
    Yes, you have zoom far in to see the difference. At least I do. Some have mentioned secondary color correction being hard, but I'm curious to see proof.

    Don't get me wrong, I would love to shoot only in RAW. It seems everything the manufacturer does to the image I would rather them not. But I'm waiting for memory cards to get to be around $30-50 for a terabyte (1 hour of 4K RAW).




    Best set of videos I've seen explaining DCT, http://www.dvxuser.com/V6/showthread...n-Really-Works

    Approachable explanation of Wavelets from the guy who wrote Cineform, https://gopro.github.io/cineform-sdk/ (see Wavelet Transforms).
    ProRes raw is the step in the right direction where I think it will start to make more sense. Media cost is one thing but there is also the workflow needs and the performance of raw to take into consideration. I downloaded some ProRes raw material that plays back perfectly fine on a 2015 MBP. Thats really impressive and a nice future path for raw. I think when ProRes raw takes off a bit more it will make raw a more realistic workflow for a lot more people other than just film makers.


    Reply With Quote
     

Page 47 of 49 FirstFirst ... 3743444546474849 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •