Author Topic: When will lens development finally peak?  (Read 2336 times)

ColinM

  • NG Member
  • *
  • Posts: 1789
  • Herefordshire, UK
    • My Pictures
When will lens development finally peak?
« on: February 09, 2021, 07:58:23 »
I read about the S lenses and am so pleased at the progress you all report.

Meanwhile, I have a little experience of sport photography and know the challenges of shooting a moving subject from around 100m away.

So I was impressed to find I can now tell if a sportsman has shaved when the (video) camera & (zoom) lens is trying to maintain focus on him as he's running fairly fast away from it. Even having bought an better spec HD TV recently I don't remember the images captured by the pitch-side equipment being this clear a few years ago....


Ethan

  • NG Supporter
  • **
  • Posts: 208
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #1 on: February 09, 2021, 09:04:53 »
.

MFloyd

  • NG Supporter
  • **
  • Posts: 1780
  • My quest for the "perfect" speed blur
    • Adobe Portfolio
Re: When will lens development finally peak?
« Reply #2 on: February 09, 2021, 17:24:48 »
Fujinon “box” lenses range from 8 to 1000 mm or 16 to 2000 mm with their incorporated TC. Price is accordingly: around 210k USD.

https://www.fujifilm.com/products/optical_devices/tv_cine/studio_field/ua125x8be/#feature
Γνῶθι σεαυτόν

ColinM

  • NG Member
  • *
  • Posts: 1789
  • Herefordshire, UK
    • My Pictures
Re: When will lens development finally peak?
« Reply #3 on: February 09, 2021, 21:15:02 »
Interesting MFloyd, I guess you see kit like this being used at some of the motorsport events you cover?

I was impressed at the Autofocus capabilites too.
The coverage of the cricket I was watching today included accurately tracking a ball travelling at up to 140 Kph over a 20 m distance, with the cameraman up to 100m away!.

Focus was good enough that you could clearly see the stitching on the ball through most of its journey...

MFloyd

  • NG Supporter
  • **
  • Posts: 1780
  • My quest for the "perfect" speed blur
    • Adobe Portfolio
Re: When will lens development finally peak?
« Reply #4 on: February 10, 2021, 03:49:14 »
Interesting MFloyd, I guess you see kit like this being used at some of the motorsport events you cover?

I was impressed at the Autofocus capabilites too.
The coverage of the cricket I was watching today included accurately tracking a ball travelling at up to 140KPH over a 20 m distance, taken from up to 100m away!.

Focus was good enough that you could clearly see the stitching on the ball through most of its journey...

Yes. I had a close look at one of these when I was in Shanghai for the World Endurance Championship in 2019. I thought I took a picture of the “beast” but, unfortunately, no  :-[
Γνῶθι σεαυτόν

golunvolo

  • NG Supporter
  • **
  • Posts: 6742
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #5 on: February 10, 2021, 11:35:18 »
210k usd... Does it come with f mount adaptor?  ;D 8)

Akira

  • Homo jezoensis
  • NG Supporter
  • **
  • Posts: 12468
  • Tokyo, Japan
Re: When will lens development finally peak?
« Reply #6 on: February 10, 2021, 11:56:16 »
I remember reading an article about the history of zoom lenses and was impressed by the ultra high magnification zoom lenses used on the TV cameras.  They employed the "double zoom" construction where a set of zoom works in the wider range and another set, in the tele range.  They used tens of lenses.  Totally unbelievable...
"The eye is blind if the mind is absent." - Confucius

"Limitation is inspiration." - Akira

ColinM

  • NG Member
  • *
  • Posts: 1789
  • Herefordshire, UK
    • My Pictures
Re: When will lens development finally peak?
« Reply #7 on: February 10, 2021, 17:21:35 »
I ...was impressed by the ultra high magnification zoom lenses used on the TV cameras. 

Agreed Akira
MFloyd - Could you share a full-resolution image from the racetrack or elsewhere that you've taken from a similar distance & taken with your best long lens? That would be a really useful basis for comparison.

I guess the best modern high reach lens would be the 850mm with TC, but your D6 is a pretty good example of what modern "still" photographic equipment can achieve.

Autofocus
Meanwhile, I've added above a capture of the TV picture showing autofocus keeping up with a cricket ball moving at 140Kph - do we think Nikon equipment can match that?
Or do we need to spend €210k to achieve that ;)

Daniel Bliss

  • NG Member
  • *
  • Posts: 28
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #8 on: February 27, 2021, 23:42:55 »
My theory is, if certain countries, especially the US, UK and Canada, finally start to take universal broadband seriously and regulate it as the public utility it always has been rather than the limited commodity it's priced and delivered as in most places, that will speed adoption of higher bandwidth processes including modestly higher resolution TV, along with the optics needed to support that. But....I also have difficulty seeing a huge rush for 8K televisions and even 4K really only delivers meaningful improvement in my view for team sports where they're always just a bit short of resolution on individual players at 1080. Set against this is we're already facing extreme scarcity and an intense need to be more efficient in our environment (if, that is, we like the idea of going on being able to eat and breathe and drink water) and when you have Bitcoin using more electricity than the entire country of Argentina for what's no more than basically another investment hedge, it puts the cost of over investment in computing into perspective.

And as for lenses? They're already more than 8K capable, in some cases more than 12K capable. The progress of the last 20 years is incredible. What more can they do? And what more will they actually sell? Camera makers are hitting the wall for this very reason already.

ColinM

  • NG Member
  • *
  • Posts: 1789
  • Herefordshire, UK
    • My Pictures
Re: When will lens development finally peak?
« Reply #9 on: March 01, 2021, 22:38:10 »
Actually I think MFloyd summed it up very well above.

These quality optics are available at, say $200,000 +

So I don't suppose either the focus systems or the glass they use will make it down to even Pro Nikon or Canon lenses.

Ilkka Nissilä

  • NG Supporter
  • **
  • Posts: 1689
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #10 on: March 02, 2021, 14:04:11 »

I don't doubt that the resolution is there in many high-quality lenses, but maintaining focus on a moving subject at the level of individual pixels is quite challenging, and arguably there is plenty of room for improvement in this area.

I was just watching a documentary "Epic Yellowstone"; there was a scene with an osprey fishing and then another bird was trying to steal the fish from it. It was very dramatic. However, the first few seconds of the osprey catching the fish were clearly out of focus, illustrating the difficulty of maintaining consistent focus across a sequence of events.

I can't even imagine how much time must have been taken from the videographer to catch this kind of a sequence of events.

The Fujinon lens has to solve a different problem compared to Nikon Z lenses as the 2/3" sensor is quite small (11 mm diagonal) but the lens's zoom range is, on the other hand, very large.

I personally think that 4K is plenty of resolution for video and it may not be all that useful to increase resolution further. Also as you note the consumption of energy can be an issue and it may be that companies have to take that into account in planning broadband and video standards. High resolution video may not look all that great if the bandwidth is very narrow. When I got my 50" 4K TV, at first there was no appreciable difference in content (FullHD vs. 4K), lately there are now more sources for genuine 4K content but it's not like the difference between 4K and FullHD is as great as FullHD vs. SD or even analog television. For 4K content to really shine, it needs sufficient bandwidth. And at some point there may be strict regulations on how much electrical power can be used for entertainment. For 8K or 12K, it may be difficult to sustain it at a level where the details can really be appreciated.

When watching movies, I think the story and other aspects of the visual presentation (such as mood, lighting etc.) are more important than fine detail.

I agree that in team sports, high resolution has benefits, but how high does it really need to be? Is 4K not sufficient? At least in Finland there are no broadcast 4K channels.

When viewing (still) photographs, one can spend quite a lot of time investigating the image and getting lost in the details in a large print. This way, high resolution is more beneficial in stills than in video. However, even for still photographs, one can argue that large prints most commonly are viewed from a bit further distance than small prints or screens, so the resolution needed is not growing linearly with presentation size. I would argue that the advantage of increasing resolution is mostly now to eliminate moire and aliasing without the use of anti-aliasing filters, thus getting closer to the full spatial resolving ability of the lens recorded in the image without artifacts.

Of course, some have the desire for infinite zooming, but I would argue it's not necessary, it's just a desire. Often the substantial content of the image can be communicated with less resolution. So it may be that the manufacturers want to increase resolution to sell more products, but at some point the customer interest is likely to wane.

Jack Dahlgren

  • NG Member
  • *
  • Posts: 1528
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #11 on: March 02, 2021, 17:04:09 »

I personally think that 4K is plenty of resolution for video and it may not be all that useful to increase resolution further. Also as you note the consumption of energy can be an issue and it may be that companies have to take that into account in planning broadband and video standards. High resolution video may not look all that great if the bandwidth is very narrow. When I got my 50" 4K TV, at first there was no appreciable difference in content (FullHD vs. 4K), lately there are now more sources for genuine 4K content but it's not like the difference between 4K and FullHD is as great as FullHD vs. SD or even analog television. For 4K content to really shine, it needs sufficient bandwidth. And at some point there may be strict regulations on how much electrical power can be used for entertainment. For 8K or 12K, it may be difficult to sustain it at a level where the details can really be appreciated.


Two things:

I think capturing at 5k or above is common even for 4k distribution as it gives a bit of room for reframing.

Larger file sizes do take more computing power to process, but transmitting data is not a big part of energy consumption. Energy consumption scales to first order with physical size of display as it is the conversion of data to light which uses the most power. The amount of energy needed to send data over fiber is very low in comparison and would easily be eclipsed by things like sealing a window.

Ilkka Nissilä

  • NG Supporter
  • **
  • Posts: 1689
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #12 on: March 02, 2021, 17:34:06 »
Two things:

I think capturing at 5k or above is common even for 4k distribution as it gives a bit of room for reframing.

Larger file sizes do take more computing power to process, but transmitting data is not a big part of energy consumption. Energy consumption scales to first order with physical size of display as it is the conversion of data to light which uses the most power. The amount of energy needed to send data over fiber is very low in comparison and would easily be eclipsed by things like sealing a window.

I'm not an expert in this but I did a bit of googling and found estimates that about 5% of the world's electricity is consumed by the internet and streaming video constitutes about 60% of the transmitted data. Streaming video is estimated to be responsible for 1% of global CO2 emissions. I would argue that that's a major issue.

Now, if you are saying that the video resolution is in itself not a significant factor in this energy consumption, that could be the case. However, to see this increased resolution, internet companies can be motivated to build new hardware with greater bandwidth and consumers can be motivated to buy new, larger screens before the end of life of their existing equipment to actually see this new detail. In this way the energy and natural resource consumption can increase.

I am not sure how the window sealing is related to the discussion? Of course, energy needs to be saved in all areas of human activity.

Jack Dahlgren

  • NG Member
  • *
  • Posts: 1528
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #13 on: March 02, 2021, 20:02:11 »
I'm not an expert in this but I did a bit of googling and found estimates that about 5% of the world's electricity is consumed by the internet and streaming video constitutes about 60% of the transmitted data. Streaming video is estimated to be responsible for 1% of global CO2 emissions. I would argue that that's a major issue.

Now, if you are saying that the video resolution is in itself not a significant factor in this energy consumption, that could be the case. However, to see this increased resolution, internet companies can be motivated to build new hardware with greater bandwidth and consumers can be motivated to buy new, larger screens before the end of life of their existing equipment to actually see this new detail. In this way the energy and natural resource consumption can increase.

I am not sure how the window sealing is related to the discussion? Of course, energy needs to be saved in all areas of human activity.

The "Internet" is a broad term and is more than just streaming. The transmission of data takes very little power so it would be inaccurate to equate the percentage of streaming video to percentage of energy used. I think it is a difficult thing to sort out the environmental costs. In the past, people used to drive to video rental stores to pick up a movie, or sit in a large airconditioned theatre. These consume an order of magnitude more energy than streaming to a local device. If streaming video (such as conference calls) removes the need for air-travel, then again the reductions are clear. Making some simple assumptions a one hour flight uses the same amount of energy per user as two years of video conferencing.

The point of window sealing is to give an order of magnitude about how much energy is being expended. My contention is that other simple things that can be done are more impactful than restricting resolution on digital devices.

Ilkka Nissilä

  • NG Supporter
  • **
  • Posts: 1689
  • You ARE NikonGear
Re: When will lens development finally peak?
« Reply #14 on: March 03, 2021, 10:20:25 »
The "Internet" is a broad term and is more than just streaming. The transmission of data takes very little power so it would be inaccurate to equate the percentage of streaming video to percentage of energy used.

I don't know how these analysis are done, here is one:

https://www.iea.org/commentaries/the-carbon-footprint-of-streaming-video-fact-checking-the-headlines

"Based on average viewing habits, my updated analysis shows that viewing devices account for the majority of energy use (72%), followed by data transmission (23%) and data centres (5%). "

https://www.interdigital.com/white_papers/the-sustainable-future-of-video-entertainment

"An 8K TV uses more than twice as much electricity than a 4K TV, which shows in energy bills; however, users are unaware  that  this  accounts for 108gCO2e  per  hour of  emissions,  2.6  times higher than the equivalent 4K set. To put this into context, for the expected 30 million  8K  TVs  that  will  be  installed  by  2023,  energy  consumption for  video streaming will  be  50%  higher  than  343  million  tablets used worldwide. As consumer engagement with sustainability increases, these choices will attract greater scrutiny."

A lot of streaming content is no doubt viewed on smaller devices. But then I have to really ask whether the increased resolution gives sufficient benefits to justify the costs.


Quote
I think it is a difficult thing to sort out the environmental costs. In the past, people used to drive to video rental stores to pick up a movie, or sit in a large airconditioned theatre.

Yes, but people did this infrequently whereas today many people "binge-watch" series on streaming services (i.e. watch the whole season in one go, for example). The increased availability of content has lead to increased consumption. This probably has disadvantages to health as well.

Quote
If streaming video (such as conference calls) removes the need for air-travel, then again the reductions are clear.

Sure, video conferencing is a good idea for long-distance communications. However, these applications use very low-quality video. Zoom transmits about 1GB per hour for video while Netflix uses 25 Mbps for 4K (11 GB/h), and I would imagine 8K would use even more than that. 4K optical discs have 82-128 Mbps bit rates (37-58 GB per hour), and if the bandwidth were available, people would likely choose to have this higher quality for streaming as well. From what I can see, the perceived image quality in video is more related to the bandwidth used than resolution, but nevertheless resolution seems to have great marketing value.

The fact that people do video conferencing despite the poor quality in my mind suggests that most people simply don't care a lot about the quality of the video, they are more interested in having a quality discussion with friends and colleagues.

Quote
Making some simple assumptions a one hour flight uses the same amount of energy per user as two years of video conferencing.

Again, I am not against video conferencing. I am merely questioning whether increased video quality (from 4K upwards) in entertainment is really beneficial and if it becomes widely used, it might increase the environmental costs and not really provide much in terms of increased substance. I've read is that today computer-generated images in movies are generated at 2K resolution and not 4K or 8K, and that the most commonly used movie cameras in Hollywood produce 2.6K files. Rendering computer-generated images at higher resolutions than 2K would apparently be too slow and/or costly. I really see no one complaining about the 2K CGI in movies. So why does the original video footage now at the consumer level have to be shot at 8K? I just don't get it.

Quote
The point of window sealing is to give an order of magnitude about how much energy is being expended. My contention is that other simple things that can be done are more impactful than restricting resolution on digital devices.

I'm not interested in "restricting" resolution of digital devices (e.g., by a regulation) but I am questioning whether it makes sense to pay for hardware to allow 8K video acquisition at the camera and lens level, then the additional processing hardware and storage needed to make editing and storage possible and practical, and finally to distribute the content at higher resolutions (and for consumers to buy wall-sized screens so that they can see the details). I'm questioning whether this added expense and effort will actually provide benefits in proportional to the cost (including monetary cost, extra time used, and environmental cost)̣ and whether instead we should focus efforts somewhere else. For example, making more interesting content rather than just higher resolution. Already resolution and image quality of video is very high.