Jump to content
ntblowz

Sharp's new 8K M43 camera

Recommended Posts

EOSHD Pro Color for Sony cameras EOSHD Pro LOG for Sony CamerasEOSHD C-LOG and Film Profiles for All Canon DSLRs

Can some help me with the math and determine what 8k 10bit 4:2:0 therorically downsamples to?

4k 12 bit 4:4:4?

4k 12 bit 4:2:2?

4k 12 bit 4:2:0?

4k 10 bit 4:4:4?

4k 10 bit 4:2:2?

4k 10 bit 4:2:0?

 

Thanks!

Share this post


Link to post
Share on other sites

@majoraxis 8k 4:2:0 has quarter resolution in the chroma channel, so it is 4k 4:4:4 with oversampled luma. That oversampling provides pseudo-12 bit, but only for luma channel. So it is 4k 4:4:4 10 bit with a nice oversampling on luma.

Similarly, you can get 4k 4:4:4 12 bit, but the chroma channel will only have 10 bits worth of information at the maximum, while luma would retain a bit more information.

Share this post


Link to post
Share on other sites
3 hours ago, KnightsFan said:

@majoraxis So it is 4k 4:4:4 10 bit with a nice oversampling on luma.

Has anyone actually proven this to be useful in practice? I remember when the GH4 (or maybe NX1) first came out a bunch of people were talking about how amazing it will be to oversample the 4K into wonderful HD with a greater bit depth and color space. I expected to see all sorts of amazing footage coming from the added image quality. But, I never actually saw anyone doing it. Is this all theoretical?

Share this post


Link to post
Share on other sites
1 hour ago, MurtlandPhoto said:

Has anyone actually proven this to be useful in practice? I remember when the GH4 (or maybe NX1) first came out a bunch of people were talking about how amazing it will be to oversample the 4K into wonderful HD with a greater bit depth and color space. I expected to see all sorts of amazing footage coming from the added image quality. But, I never actually saw anyone doing it. Is this all theoretical?

It's not theoretical. Check out 4k footage from the NX1 vs. the 1080p footage from the same camera. Downscale the 4k footage to 1080, and compare them at that HD resolution. It's a night and day difference, from the same sensor, processor, and codec. You can do a similar comparison on most other cameras as well.

Whether it's "useful in practice" depends on whether you're talking useful to the general public's enjoyment of a compressed YouTube file, or useful to a cinemtographer's critical eye looking at a high bitrate master.

And just to be clear, downscaling in post doesn't gain quality. It loses quality. The original 8k 4:2:0 file has better information fidelity and more spatial resolution than a 4k 4:4:4 file downscaled from the 8k original. The only claim is that downscaling to 4k 4:4:4 will retain more information than downscaling to 4k 4:2:0.

Share this post


Link to post
Share on other sites
3 hours ago, MurtlandPhoto said:

Has anyone actually proven this to be useful in practice? I remember when the GH4 (or maybe NX1) first came out a bunch of people were talking about how amazing it will be to oversample the 4K into wonderful HD with a greater bit depth and color space. I expected to see all sorts of amazing footage coming from the added image quality. But, I never actually saw anyone doing it. Is this all theoretical?

This question comes up a lot.  Here's an answer I prepared earlier.....  please excuse the first paragraph (it was directed at the post I was replying to, not your post).

On 2/24/2019 at 1:16 PM, kye said:

This is a tricky subject, but you have nailed it with your comment "seems to be a lot of opinions".  Just like everything else out there, if something is engineering or science, there will be a lot of opinions, and almost be definition they will all be WRONG.  People who have OPINIONS about engineering or science are people that don't understand FACTS.  I'm all for having opinions, we can talk about who likes what colour science, lighting design preferences, lens aesthetics or if someone is a good actor, but anyone who has an opinion about how many pixels are in the UHD specification is just stupid.  This is the same thing.

There is a huge level of knowledge about how to get accuracy beyond a certain bit-depth when talking about audio, as properly recorded and processed 16-bit audio can have better signal-to-noise ratios than is mathematically possible because of a technique called dithering which works by adding a very specific type of noise to the signal.  

https://en.wikipedia.org/wiki/Dither

Fortunately, ISO noise on high-quality 4K cameras is a relatively good version of that noise, so we can get a lot of the benefits.

Downscaling from 4K to 1080 also involves oversampling which when combined with dither can extract the extra bit depth and eliminate the noise that was added.

https://en.wikipedia.org/wiki/Oversampling

There is an audio format called SACD which uses a type of digital signal called DSD, which is a 1-bit (yes, the bit depth is one bit!) at 2.8224 MHz, and because of its clever use of noise and processing, can have signal-to-noise rations of up to 120dB, which would require a 20-bit signal from a traditional codec, but because it is oversampling (in a big way) this effect can be achieved.  Getting 20-bit from 1-bit is only possible because DSD has about 64x the sampling rate compared to 44kHz audio.

https://en.wikipedia.org/wiki/Super_Audio_CD

If DSD gets 19 extra bits from a 64x oversampling, then it shouldn't be impossible to do a similar thing with video and get an extra bits from resolution oversampling.

However, and this is a key part of the picture, you will only get perfect 10-bit 4:4:4 1080p from 8-bit 4K footage if that 4K footage is RAW and the noise is perfect.  Any variation in de-bayering, compression or any other processing that is applied in between that data coming off the sensor and the downscale will have a damaging effect on the final result, and this is where reality differs from theory, and it the overall quality will be different depending on the camera, codec, bitrate, subject matter, and probably other things.

If none of that made sense, then here's a TLDR approximation - adding noise to 8-bit helps with banding similarly to why adding noise to your footage helps with YT colour banding.  The mechanism is very different, but the effect is broadly similar.

Anyway, let's put this to bed and go back to talking about cameras 😆😆😆

In terms of people actually doing it, everyone who shoots 4K and publishes 1080 is doing it.

Share this post


Link to post
Share on other sites

Yes, that 5.5 almost bezeless FLIP-OUT smartphone-alike screen is GORGEOUS and I dont undestand why other camera manufactures dont use those kind of screens. 

Three key things missing for me in this camera - IBIS, high frame rates and AFC. But it is still an acheivment and I am glad that Sharp joind MFT game.

Also impressed by the sensor lowlight abilities. 

 

 

Share this post


Link to post
Share on other sites

Having seen how effective the stablisation is on an Insta360 One X when producing flat content from overcapture, I don't have any concerns really about this camera, or any 8K camera for that matter, not having IBIS.

Unless you are delivering in 8K then it gives you plenty of latitude for post capture software stabilisation. With some of the side effects I've seen from IBIS, which are then baked in to the footage of course, and the advances in software stabilisation its arguably a better approcach anyway. 

 

 

Share this post


Link to post
Share on other sites
18 minutes ago, BTM_Pix said:

Having seen how effective the stablisation is on an Insta360 One X when producing flat content from overcapture, I don't have any concerns really about this camera, or any 8K camera for that matter, not having IBIS.

Unless you are delivering in 8K then it gives you plenty of latitude for post capture software stabilisation. With some of the side effects I've seen from IBIS, which are then baked in to the footage of course, and the advances in software stabilisation its arguably a better approcach anyway. 

 

 

..and that is just 5.7K. With 8K can have 6K stabilized. Easily..

People misunderstood resolution, it is needed for more than one reasons, and not native (8K-6K-4K) delivery.

Share this post


Link to post
Share on other sites
1 hour ago, Kisaha said:

..and that is just 5.7K. With 8K can have 6K stabilized. Easily..

People misunderstood resolution, it is needed for more than one reasons, and not native (8K-6K-4K) delivery.

Absolutely.

1080p will be the target format for YouTube for a good while yet and with the stabilisation and re-framing possibilities, this camera has the potential to be the best camera for a sizeable number vloggers.

The faux slider move function on Panasonic 4K cameras is pretty decent as is the faux multicam of the MEVO camera and both of those features will be greatly enhanced by the extra resolution of 8K.

The camera also has decent audio options and of course that huge articulating screen.

For someone working on their own who needs to shoot content fast with the ability to add production value in post, this camera has the potential to be an absolute powerhouse.

Its interesting to see a lot of the reactions to this camera as dismissing it as a bit of novelty if not a flat out joke and I'm sure a lot of that is related to it coming from a manufacturer that is not one of the usual suspects.

I think if people look beyond that, which they should because Sharp have not only been around the block a few times but are owned by Foxconn, and consider the possibilities that it offers then they might well find that this has the potential to be a far more signficant product than it is being given credit for.

 

Share this post


Link to post
Share on other sites

As an FYI for anyone who hasn't seen the Mevo.

It is let down really by having too wide a lens (and small sensor) so with a 4K capture resolution you are pushing it a bit on punch ins but it is not difficult to imagine how those issues will be mitigated with the Sharp camera 

 

Share this post


Link to post
Share on other sites
2 hours ago, BTM_Pix said:

Having seen how effective the stablisation is on an Insta360 One X when producing flat content from overcapture, I don't have any concerns really about this camera, or any 8K camera for that matter, not having IBIS.

Unless you are delivering in 8K then it gives you plenty of latitude for post capture software stabilisation. With some of the side effects I've seen from IBIS, which are then baked in to the footage of course, and the advances in software stabilisation its arguably a better approcach anyway. 

Stabilisation in post only works with very short shutter speeds, so goodbye 180 shutter!  Sure, you can stabilise slow shutter clips, but the motion-blur that occurs during the frame makes the image just have instantaneous blur turrets syndrome, which is probably more distracting than mild camera movement.  If you want to use a slower shutter then you need OIS or IBIS to stabilise during each frame being exposed and then you can stabilise in post.

That's my approach to how I shoot handheld with my GH5 and stabilising in Resolve.

Share this post


Link to post
Share on other sites
1 hour ago, kye said:

Stabilisation in post only works with very short shutter speeds, so goodbye 180 shutter!  Sure, you can stabilise slow shutter clips, but the motion-blur that occurs during the frame makes the image just have instantaneous blur turrets syndrome, which is probably more distracting than mild camera movement.  If you want to use a slower shutter then you need OIS or IBIS to stabilise during each frame being exposed and then you can stabilise in post.

That's my approach to how I shoot handheld with my GH5 and stabilising in Resolve.

I was taking using lenses with OIS as a given to be honest.  I actually thought I'd put that in but my old brain is leaking like a sieve these days!

I also think the market I'm talking about is also less sensitive to sticking to 180 degree shutter rules when you look at their output where its clear that many of them are shooting in aperture priority without ND, not to mention the abundance of slooooow mooooooo in their B roll ;). Where this camera scores in that latter regard is that for many, though not all I have to stress, the slow mo B roll is very often a crutch to enable them to spread the coverage they shoot a bit thinner whereas I think re-framing, particularly the dynamic type, offers more interesting alternatives to that approach.

Just to be clear and go back to what I said a few posts ago (or maybe even in the other thread !) about this camera, I think its strength lies in what I would loosely class as 'video' work rather than the even looser term 'cinematic'. 

Its very much in the vein of the shoot first frame later scenario that the 360 cameras have given us a glimpse of but just that this does it with fewer compromises that are inherent in those cameras in terms of image quality and fisheye lenses !

I think that's what will make it a very viable option for people who have seen that potential with the 360 cameras but want more control over what's going into the camera.

Of course, I'm saying this without knowing the price so if its over £3K then all bets are off !

Realistically, it should actually be half that price considering who is behind it but when Panasonic are still getting away with marginal re-hashes like the G91 for £1K then I suspect it will closer to £3K than £2K.

Share this post


Link to post
Share on other sites
3 minutes ago, BTM_Pix said:

Of course, I'm saying this without knowing the price so if its over £3K then all bets are off !

I think I heard somewhere it’s going to be like $4-5k. Could be making it up though.

Share this post


Link to post
Share on other sites
4 minutes ago, Anaconda_ said:

I think I heard somewhere it’s going to be like $4-5k. Could be making it up though.

In which case, I withdraw all of my carefully considered advocacy of it and declare it to be an overpriced toy for rich people to shoot videos of their cats  purely to have something/anything to show on their equally overpriced 8K televisions ;) 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...