Jump to content

Beritar

Members
  • Posts

    206
  • Joined

  • Last visited

Posts posted by Beritar

  1. I'm more and more interested by the Nikon Z system.

    I'm a hardcore Panasonic GH and S user but I'm disappointed by the S5II and the S5II(X), they improved the AF and the IBIS a lot, but the image quality in video is not there anymore for me.
    The only advantage they have compared to other system is the IBIS now.
    Of course the 24-70mm, 70-200mm and 50mm S Pro are great lenses, but where are the 24,35,85,135 S Pro ? And I don't even speak about f1.2 lenses. 

  2. On 1/6/2024 at 6:01 PM, MrSMW said:

    I haven’t found it huge or even that significant, but the output from the S1H is definitely ‘nicer’ and more ha, ‘cinematic’.

    Has to be the combo of that different sensor and the OLPF.

    I have been giving it some thought just how I might use my pair this year and had settled on S1H + battery grip + 70-200 on tripod with S5ii + 28-70 on roaming duty, but I’m thinking now I might just pick up a battery grip for the S5ii, take the grip off the S1H and then swap the lenses around so the S1H becomes my run & gun.

    6k 30p for the S1H and the S5ii can do 4k 60p for the long static stuff and tracking shots. I could and even perhaps should, use 4k or even 6k 30p for the static stuff, but I’d forget to switch and find that every time I wanted some slo mo, I had no suitable footage, heat of the moment and all that!

    Do you shoot with the standard profiles or in V-Log ? A lot of people don't really see the difference in V-Log, but the standard profiles are really much worse, especially in 4K or C4K, a very bad mix of over-sharpening and chroma noise reduction.

    This is why I use most of the time 6K or 6K Open gate V-Log on the S5II. But on these resolutions, the IBIS is much worse than on 4K and C4K... Walking while recording is very difficult because the cameras makes more "jumps".
     

  3. 7 hours ago, kye said:

    IDK, it doesn't look too bad to me - especially considering the extreme level of pixel peeing going on here:

    S1vS5ii_1.2.1.thumb.jpg.ecd3f8e66185176c111b42a167f5bcc1.jpg

    S1vS5ii_1.1.1.thumb.jpg.21584cf783c451c10ffe59a0c8fe04c2.jpg

    By the time you're viewing this at 1:1 on a UHD display after you've exported the footage, I'd seriously doubt there's a meaningful difference to it.

    When you apply a blur to the footage, the first thing that it does is to reverse any sharpening, then when you apply more it starts pushing it into what people might think of as soft.

    The mathematics of blurring and sharpening are opposite.  Blurring is literally "un-sharpening".  To get more technical, both operations adjust the frequency response of the image by adjusting the high-frequencies in the image.

    Think of it like a WB, if the image is too warm then you make it more blue, but the image doesn't instantly become blue.  The reason it doesn't become blue instantly is because it has to cancel out the warm tint first, and only after that is cancelled out will the image become blue.  Applying a blur is like doing a sharpening balance, where you un-sharpen to bring the image back to a more neutral place.


    Each time I apply blur to my footage on the S5II, it becomes unatural, some textures look good and some don't. The sharpness is not the only issue, like I said. The S5II/X have more spatial chroma noise reduction, leading to loss of colour detail, like in skin texture and foliage, about a value of 20~25 in Resolve is required to match the S1/S5 chroma detail to S5iiX.

    But yes some people are prefectly fine with the internal output of the S5II, especially if they don't crop a lot inside their videos, but to my eyes, the S5II internal recording is really not as good as the S1, S5 and S1H.
    I am not even talking about the standard profiles in 4K or C4K, one of the worst image I've seen between every Panasonic cameras I owned (almost all since the GH4), the difference with the S1,S5,S1H is huge, very excessive sharpening and smoothed out details (I think I've posted some videos about the issue).

  4. 14 hours ago, MrSMW said:

    My S1H is better at everything over my S5ii, stills and video, except outright AF capability and size/weight, the latter of which is not an issue to me.

    Yes the image is much nicer in my opinion. But outside of the AF, the IBIS of the S5II is also better, I can almost walk with the S5II with my 85mm S, you really can't on the S1 and S1H. And some features are great on the S5II too, like the zoom while recording.
    I hope the new S1II/S1HII will get the best of the old and new Gen.

  5. 2 hours ago, John Matthews said:

    Is this H.264 or H.265 footage? On Panasonic, H.265 messes more with the image IMO. I remember the GH6 having crushed blacks in H.265.

    You can't use H264 with 5,9K/6K/Open Gate.
    I don't see any crushed black on the GH6 between H265 or H264, the crushed black is only with Prores, it's because the levels are not set correctly.

  6. 1 hour ago, kye said:

    Fascinating comparison.

    How do they compare if you blur the S5 to match the sharpness of the S1?

    I read a thread on Reddit some time ago asking if people should use sharpening, and maybe half of the replies were people saying that most of the time they actually add a very slight blur to the footage (and the context of the thread was REDRAW, so no in-camera sharpening at all) and that adding any sharpening runs the risk of looking like video.

    I just about laughed out loud, because deliberately blurring your footage is against the religion of most YT film-makers, but secretly the pros are doing the opposite.

    After working out that careful softening of footage is a valid and desirable part of colour grading, I became far less concerned when 4K footage was sharpened in-camera.  I suspect the image would be better without the sharpening, but if it is sharpened and then compressed in-camera, and then you're softening it in post then you're actually softening the edges of the compression artefacts too, which should make the image better and not worse.  

    TLDR; the pros subtly blur footage so don't be afraid of it; and if a camera is too sharp then try blurring it and only evaluate the image after you've created the desired sharpness.

    If you blur the S5II, it will not get the same level of details of the S1.
    And vice versa, adding some sharpening on the S1 footage still make it nicer than the S5II. The S5II shows really harsher edges + more spatial chroma noise reduction, nothing comparable to when you add in camera sharpening in the S1 or when you add sharpening in post in Premiere or Resolve. I don't understand why Panasonic messed up their internal video engine like this, they were so close to perfect before.
    I don't want my image to be soft, I want details, but organic details, unprocessed. 
    If I want a soft image, I still can shot in 1080P and upscale it.
    This is where I'm lost when I see some people buying 6K or 8K camera to finally destroy most of the fine details to look like 1080P footage

    I think the S1 has a really good balance between organic and sharp. For me it is still one of the most underated hybrid camera.
    The case of cameras like the S1H are different, they don't blur anything digitally, and while they make the image softer for both photo and video, they are most of the time free from moiré while still having a lot of details.

  7. 18 minutes ago, Al Dolega said:

    The non-log profiles in my S5IIX do have a bit of that gross cellphone-like edge sharpening. Some people have blamed this on the addition of PDAF, citing Panasonic's past statements about PDAF affecting picture quality, but I don't see how that could affect the picture globally, I would think that would only affect the phase pixels.

    Seems to me that this could only have to do with the new processing that they touted at release, as part of the "new" L2 partnership with Leica. Which if Leica is actually responsible for grossness, is surprising given their historical tradition for natural image quality.

    I've only used V-Log for years now so luckily it doesn't really affect me, and I haven't noticed a difference in my S5IIX footage.

    The B-Raw implementation on the new cameras is better, so there is some small compensation there.

    Agreed, it is not because PDAF because the external BRAW is flawless.

    However, unlike you, I see the difference with the first S cameras almost each time, maybe  because I use big screen with high resolution, I also crop a lot inside my 6K video to make some effects.
    Yes V-log is less impacted than the standard profiles like you can see in the comparison, but for me it still not as good, it also has more spatial chroma noise reduction even in V-log. 

  8. For those who are interested by the difference of the fine details rendering (internal recording) between the S1 and the S5II in 5,9K.

    V-Log, ISO 4000, sharpness and NR at 0, internal recording, 10 bits, same lens, same settings :

    V-Log.thumb.jpg.7845c0d17fea509dedee406a63faf60f.jpg

    The S1 has at least the same level of detail but looks more organic. The S5II on the contrary, has rougher edges. 


    Natural, ISO 1250, sharpness and NR at -5, internal recording, 10 bits, same lens, same settings :

    Natural.thumb.jpg.99f4a0f5b57e67462bfbf9b237866162.jpg

    This time the difference is much worse with the standard profiles, it looks like smartphone footage, of course there are more details than on smartphone (even with 8K smartphone), but they are really over-sharpened.

    On the S5II, I never use 4K and C4K because they are even worse, especially with the standard profiles.
    But using the 5,9K/6K or Open Gate mode decrease the IBIS performance, it's very noticeable while walking (of course ambassasors never test the IBIS with these modes).
    I really hope the rumored S1II and S1HII will improve the video engine of the S5II.


     

  9. 27 minutes ago, John Matthews said:

    My god, how many systems do you actually have? That's quite a collection.

    Not so much.
    My wife uses Sony cameras, I'm not a fan of them but I like some FE lenses so I use them occasionnaly ...
    Otherwise, I only use M43 and Panasonic L mount (I also have my good old NX1 and NX500).

  10. 5 minutes ago, John Matthews said:

    I've always said those Olympus pro lenses weren't really in the spirit of M43, but that doesn't mean they aren't great lenses. I have the 17mm pro. When you look at the final image at any reasonable magnification, they look more than great. As for the 17mm, it's difficult, if not impossible, to get that sharp, shallow DOF on a wide lens, that 17mm gets you close as long as the subject isn't that far away.

    Yes, on M43 it's hard to get the same look than a FF 35mm f1.4 or f1.2.
    I have the 35mm GM for that, awesome lens, ultra sharp wide open with nice bokeh, and not so large and heavy. It's one of the best lens I ever used. I would love to put this lens on my Panasonic S cameras ...

  11. 12 hours ago, John Matthews said:

    There were some other important point of the lenses + bodies I put up earlier and would probably favor Panasonic L mount:

    • Breathing correction
    • Consistent colors
    • Size for gimbals
    • Consistent AF motors
    • Consistent fall-off
    • Consistent look

    Actually, I really only see Olympus PRO lenses being similar to the Panasonic FF lenses. Maybe I'm wrong? It would seem Panasonic is developing from the get-go whereas others are developing by answering the question: "where's the hole?" E-mount and M43 is kind of understandable because they've been around for so long. Cinema glass seems similar to the Panasonic method as they need them to quickly swap on follow focus systems.

     

    The 25mm and 45mm Oly Pro lenses are even heavier than the 50mm and 85mm f1.8 S. But yes both the Oly share similar size and weight.

    Technically, you can say they are close to f2.5 for DOF and Bokeh, they are also less sharp wide open than the S 50mm and 85mm Panasonic FF. But their bokeh and rendering are great, and we can't get the same IBIS quality, the same rolling shutter and the 4K 120fps when using the S lenses on the S5II. 
     

    P1241076.jpg

  12. 12 hours ago, PannySVHS said:

    What about a 50mm 1.8 bargain lens on the S1? @Beritar  That should be much smaller. The Sigma 45mm 2.8 is pretty small but is a relevant amount narrower regarding fov than a 40mm. No real equivalent to the nice 20mm 1.7 in the FF world other than the Voigtlaender 40mm F2 in M-mount or a Konica with a long flange and adapter.

    I love the fov coming from a 40mm lens on a FF camera. A native 40mm 2.8 pancake would be sweet, with Af and a clutch for hard stops when in manual focus. The G9 II in S5II body is dwarfed by the gigantic S1.😂 Oly 35RC has a 42.5mm F2.8 lens btw. I loved that lens and focal length and that camera.:)

    For sure, the 50mm f1.8 S is much smaller with the S5 or S5II. Very sharp lens, almost clinical, with good bokeh at short to medium distance.
    But the G9II with the PL 25mm is still nicer to use (smaller and lighter) and has IMHO better character :

    P1004398.thumb.jpg.44cc6bf1238dbc105c2d77ff63e09366.jpg

    And just for fun, a comparison with the little 20mm pancake 😅 :

    P1004400.thumb.jpg.f15383227c73d4b653b2e06fd55e98b0.jpg

  13. 2 hours ago, John Matthews said:

    Which lens f/1.8-ish + small-ish telephoto line-up looks best? Which would you rather shoot with?55209526_Screenshot2023-11-25at18_35_06.thumb.png.216a9cc620b0a6549fa0722ccd92fe68.png

    1015099138_Screenshot2023-11-25at18_33_24.thumb.png.0ff65ac3230f77f50ae2f46296a6688f.png

    2066590441_Screenshot2023-11-25at18_44_44.thumb.png.363e21637260aca2e8cd104a08dfb885.png

    In doing this exercise, I realize that it would seem Panasonic M43 and the Sony lenses were seemingly designed rather independently whereas Panasonic L was really designed all at once.

    Always amazed by the size and weight of the 25mm f1.4 (or 20mm f1.7) and 35-100mm f2.8. When I compare them to my 50mm and 70-200mm (f2.8) S Pro, I think I will not sell my M4/3 stuff anytime soon.
     

    P1241075.thumb.jpg.bac44ba14c5d9bf2d944e102bd680f26.jpg

    There are some alternatives to the 25mm PL for full frame, I tried the 45mm f2.8 Sigma for my S1/S5III (very similar for dof, bokeh etc) but I prefer to use the PL, better rolling shutter and stabilization on the G9II.
    I really can't see an alternative to the 20mm f1.7 and the 35-100mm f2.8 though.

  14. 5 hours ago, IronFilm said:

    I wonder if Panasonic would be crazy enough to bring out a GH5 mk3, that is simply identical to the GH5 mk1 / mk2, but with DPAF added! Especially if they could hold the price the same, or even better drop it slightly to fall under the magical US$1K mark. Could be a smash hit then. 

    (although, I would dream that they make one more addition to the GH5 mk3: add timecode support! At least like the GH5S has)

    The GH7 can get the kitchen sink at it, but let the GH5mk3 carry on all that is great with the GH5, but with PDAF

    A GH5 with the colors of the GH6 and the AF of the G9II for about $1K would be perfect for me.

  15. 15 hours ago, PannySVHS said:

    I don't care about S5II/X on my own behalf as I am not intending to trade it in for any of my stuff I am going to have to sell. I got too much unused equilment, so a complimentary camera instead for my S1H will be cool. At the moment that would still be my S1, which I would sell. A G9 would offer additional features to find out about, especially Dual Gain, Pdaf and punsh-in preview during recording. Wanna test that Towerjazz magic and new colour!:) I have seen lowlight comparisons with the GH5s where it holds onto resolution and colours better than the latter. I just wouldn't buy into mambojambo affected textures. How could they mess up on that. Flagship cameras, at least from Panasonic, should not have unnecessary image nasties like that. If a GH5 (ll) and G9 do better in that regard it is a step back, which is unnatural by the logic of a succeeding model of the product line.

    So, Panasonic, do your GH6 and G9II a favour and make them the much better cameras they potentially akready are. Turn off the mambojambo.

    There is a benefit though. Much less moiré than the GH5 and G9, also the footage can look more cinematic without doing much in post because some fine details are smoother.

    The GH6 and G9II also use more noise/chroma noise reduction with most profiles than the first GH5 and G9. To get ride of it, you must use the "huge" 5,7K Pro Res mode in V-Log. But the AF is so bad in 24/30fps (AF accuracy issues) that I never used it.

    In any case, to really see the difference you need to crop a lot, and the details become really nasty only in low light, I'm sure 90% of users will never see the difference, like most never seen the edge sharpening/chroma noise reduction difference between the S5II and S5/S1.
    I even think most people prefer to get slightly altered details with less noise than more preserved details with full of noise.

  16. On 11/20/2023 at 9:06 PM, PannySVHS said:

    How was rolling shutter on these babes? Iirc, they are fine up to Iso 800, from there it's mush galore. Would love to test a hacked NX500 some day. But actually I should just use the gear i own instead of buying anything new or old!😂

    I still have my good old NX1 and NX500.
    Indeed, even at ISO 200, with bad light, details fall apart, very mushy.
    Back in 2014/2015 I was really disappointed by this mushy rendering in low light compared to my GH4, I was one of the first to point out  : 

    This exactly what the S5II is doing with most picture profiles (internal), of course the S5II has much better low light performance so it's much less worse, but when I crop inside the low light videos of both my S1 and S5II, I can see more or less the same difference, the S1 keeps much better details than my S5II (V-Log is really better but details are still not as good as on the S1, mostly because of weird edge sharpening).

    About the rolling shutter, it is really bad in 4K on the NX1, one the worst I've seen, but the image was and is still gorgeous with the right light, especially with the 16-50mm and 50-150mm NX S. Iirc, the RS is close to 30ms.

    I also have the NX500 somewhere in my house, but I never really liked the video quality compared to the NX1. In good light the NX1 is full of details, the NX500 much less, probably because of the big crop. Also the NX500 has more difficulty to nail the focus in 4K, and it has no electronic stabilization, which is very handy with the little 30mm and 45mm on the NX1.
    I used the hacked firmware of the NX500 allowing the 2,5K mode, the rolling shutter was excellent, below 10ms for sure, but the image quality was not good, there was sometimes weird lines and aliasing.

  17. 5 hours ago, PannySVHS said:

    I just hope that Panasonic allows to turn off the mambo jambo with textures, what @Beritar was describing. Then it would be my S1 replacement and compliment my S1H. I am curious about dual gain colour. I would also love to have more "official" crop modes for S16, 2/3 inch, 16mm and so on, rather than a 2 x digizoom for 2/3" or pixel to pixel , which on the G9II in 4K mode is something around the 16mm format. Btw isn't Beritar a label under which Tokina and Sigma lenses have been sold, just like Vivitar, Soligor, Hanimex, etc? 🙂

    Sadly, this is just a stupid nickname 🙃.
    About the mambo jambo textures, when using BRAW on the S5II, the image is free from the weird edge sharpening, the image is great, juste like the internal S1 and S5 with their last firmware. So yes, like you I hope they will allow to turn off most of their bad processing, they surely can.

  18. 7 hours ago, kye said:

    Yes, lots of stuff is done in-camera and cannot be un-done, so it's just a case of trying all the tricks we have and seeing how far we get.

    If you do a custom WB on the G9ii and GH6 on the same scene, does it remove the differences in WB between them?

    Hard to say because I already sold the GH6. However, like the S5 and S5II, the G9II and GH6's colors are slightly different.


    But honestly the G9II's colors are really good, the best I've seen (with the GH6) from a Panasonic camera.
    From what I've read, the sensors inside these two M43 cameras come from TowerJazz, as opposed to Sony with the GH5/G9/S5II/S5/S1H/S1. 

  19. 8 hours ago, kye said:

    Just un-sharpen more.

    Unless you like the digital look?  TBH I find that if something hasn't been un-sharpened, even if it was shot with RAW video and not processed at all, I find that it is trying to shout at me over the content of the video.

    I suspect the culprit is the post-sharpening that is done by YT / streaming service.  Comparing your upload to YT shows quite a substantial difference.

    Ahah, no, of course it is on the original files, I never pixel peep on converted files.

    I don't like the digital look at all like on smartphone (when everything is oversharpened), but I don't like digital processing making the image soft or sharp artificialy either. I sometime need to keep all the fine details for my videos, I don't want they have digital sharpening, I just want the same details as a raw photo without any digital processing. If I really need a soft image looking, I think the 1080P of my GH5 with a bit of un-sharpen is perfect, no need for a 4K or 6K camera.
    If I un-sharpen a video file looking like a raw photo, as on the GH5, the result is good, because every details are there. But when I un-sharpen a file like on the GH6, it just doesn't look as good in my opinion, especially in low light, it looks more like if the image is upscaled.

    My S5II also has bad processing going on internaly compared to previous S cameras, bad edge sharpening even in V-log, making the fine details less thin, and also chroma noise reduction more or less significant depending on the profile used. Of course, most of the time and depending on your screen, you need to crop in order to see the difference.

    Digital processing like the Intelligent Detail Filter on the GH6 can certainly helps with moiré, but nothing come close to a OLPF like on the S1H for the nicest image quality. I even prefer to use my GH5 in low light compared to the GH6 below 1600 ISO, sure the GH6 is cleaner, but something just looks off about fine details rendering. Of course, downscaling the file to 1080P or viewing it on small screen doesn't make a big difference between both cameras. Again, it's really when cropping we can see that the details rendering is not the same.
    But it is not so simple, because the colors of the GH6 are really really nice, I just love their new 25MP M43 sensor, it has natural and rich tones in video (and photos of course), nicer than on my S5II IMHO. Last time I loved the colors of a camera as much was with the Samsung NX1.
    My G9II has more or less the same colors than the GH6, but on the cooler side, and more magenta overall. Nothing that can't be fixed in post though.

  20. 18 hours ago, kye said:

    Just un-sharpen in post.

    I remember a great thread on a colourist forum some months ago asking about sharpening, and the responses were that mostly people don't sharpen, and many reduce the sharpness of the image to avoid a digital look.

    I think that unsharpening might be one of those hidden things that camera fondlers would consider heresy but is widely done by the pros for high end work.  They were talking about cameras that shoot RAW too, not compressed codecs from consumer cameras.

    Yes but it is not only a problem with sharpening, both the GH6 and the G9II have a digital filter to remove moire, but the result is not so great, some details looks smoothed out, especially in low light.

    I much prefer the details of previous Panasonic cameras. On the old G9 and GH5, the fine details looked really like on a raw photo, it was very sharp but like you said it's pretty easy to un-sharpen in post. 
    But on the GH6 and G9II, the Intelligent Detail Filter just destroys fine details making some areas of the image soft and some areas very sharp.

  21. 31 minutes ago, John Matthews said:

    The 70-300 seems especially good with dual IBIS. I've seen some really decent footage. It's the only telephoto lens I'd consider. I'd much rather have it because you get an extra 100mm and only lose a stop over the 70 200 F4. That's an easy tradeoff; also, it's a bit lighter.

    I've heard good things about it terms of IQ, but apparently it's rather heavy for what it is.

    Yes, IBIS is very good on this lens, especially with the S5II and S1. Sometimes you can get it for about 800€ in Europe with cashback. It's a fine lens, even the colors are pretty nice.

    The 70-200mm is pretty heavy yes, but the Sigma is not so much lighter. The bokeh is nicer on the S Pro. And I just love how the light renders on this lens, the local contrast and colors, this is truly beautiful, it's similar to the 24-70mm S Pro and it is why I prefer the S Pro lenses over the Sigma DG DN.

  22. 6 hours ago, John Matthews said:

    Have you considered the Panasonic 70-300? I've actually heard a lot of good things about that lens, even though it's a little slow. It probably does ok at ISO 4000 indoors.

    Yes, the 70-300mm is a nice "little" zoom. I have both the 70-300mm and the 70-200mm S Pro f2.8.

    The 70-300mm is sharp across all focal lengths, maybe a bit softer at 300mm but still very good.
    The difference with the 70-200mm F4 is not so huge since the S cameras have excellent low light performances.

    The 70-200mm S Pro f2.8 is a different beast.  Yes, it's sharp, but the rendering is outstanding, maybe because of the implication of Leica (or not), one of the best lens I own with the 24-70mm and 50mm S Pro. 

×
×
  • Create New...