If you are shooting 36mp still images, you're going to be able to see a pretty significant difference when comparing cheap lenses to their pro level counterparts. Does the same hold true for video? Not really.
Why is video any different? Well unlike a still image, that you can stare at indefinitely and notice every detail, video is a bit more forgiving. Every 1/24th of a second another frame is flashing in front of your eyes. The amount of data/second that your video footage is played back at may end up effecting your sharpness more than your lens. You might have the sharpest lens in the world but your camera may record footage at a low bit rate or maybe your camera is super high end but once the footage was edited it was exported at a low bit rate.
Perhaps the biggest argument against ultra "sharp" cine lenses is that most footage these days is played back at 1080p (about 2 megapixels/frame). At 2 megapixels, it's hard to notice many details at all. Remember that no matter how cheap a 35mm lens is, it is made to produce a sharp image that is well over 20mp. Most of Canon's lenses are still working just fine on the new 50mp 5Dsr. In the video they did film at 5k but they still ended up showing the footage on a 1080p screen.
At the end of this video 3 people compare all of the footage and try to guess which lenses filmed which scenes. The professional cinematographer who had the best score was only able to guess 8 out of 18 scenes correctly.
Once again, it has been proven just how amazing and capable even the cheapest photo/video gear on the market today currently is. It also shows why a 50mm 1.8 lens is the absolute best deal for the money for any camera system. There is no reason not to own one.
Good comparison! But if these glasses are really that similar, what makes them so much more expensive?!
well but even if they produce not that many lenses and use better glass and the focus ring is much more smooth and precise, does that justify this difference in price?! I didn't miss any part! ;) But for me, this doesn't justify this hugh difference.. Maybe they should have compared also canon's 50 1.2?!
more like a lens comment. i mean it's clear that there has to be a difference between a 150 and a 15'000 glass. After all, photography and cinematography are 2 different things.
15K is actually quite cheap for cinema glass.
your average Master/ Ultra Prime costs 20+ upwards per focal length. An Angenieux Zoom starts at around 45+ and a Hawk Anamorphic at 55/60K+
yeah ok i agree with you, but what made me think is the fact that these 3 guys didn't really see any difference.. i mean 8/18?!
(not meaning to sound like a jerk. This is the industry I work in and I'm very passionate about the quality of glass we work with)
Let's also remember a few things. VERY few people actually own their own cine lenses. Production companies sure don't. Many wealthy DPs might own their own set of lenses but still wouldn't use them on every shoot because (wait for it) it might not be the right tool for the job.
So, speaking of the right tool for the job, we're talking about Feature Films and TV shows that cost millions of dollars to make. A rental company (like the one I work for) will spend the huge money on a set of lenses because they're reliable, consistent, durable and easy to repair/adjust as needed. I don't think we're expecting anyone to be buying a set of Ultraprimes (or masterprimes HA) for your DSLR documentary. There's a right tool for the right job (or at least a more correct range of tools).
Here are some of the things that the industry expects to perform better in Cine lenses:
-Sharpness (in the whole image)
-Ability to hold focus (for zooms)
-Creating a standard housing that has the same front diameter for a set so that all mounted accessories are universal
(I'd love to see what the VisFX folks think about the lens mapping on the best L-Series lens you can find.)
Cine lenses are also very easy to service and have high quality parts that *ideally* don't break as often. Since they're set up with a manual focus puller, there's not need for internal motors that can easily break under stress.
-Sharp lens markings
-PL mounts are a slightly more solid design with less play for focus shift (arguably)
-Better performance at lower apertures
But at the end of the day, let's talk about the right tool for the job and HOW the industry is using/acquiring these tools.
I think it's more about build quality and the way a lens handels that make it more expensive.
- the canon lens takes you 10 shots to nail the focus right in a shot
- the zeiss lens takes you 2 shots to nail the focus right in a shot
Lets say for example it takes you 20 minutes every time to set up this shot. Then it woud save 2 hours and 40 minutes on set. Everyone on set has to be paid those extra hours. So in the long run it would save you a lot of money buying the expensive lens.
And after shooting a whole movie, the canon lens which i owned and is great, (till i discovered i like nikon better) will probably be replaced about 4 times because it broke. The zeiss lens most likely would have survived..
I don't do Film (only photography) so i don't know for sure if i'm correct about this, but this is why i can image people buying a more expensive lens and justifying the pricing of it. It is the reason i only buy top quality lenses for my photography business..
I don't know if this is just me? But the last shots at night, you can really see the difference in the images. The "bokeh balls" In the first shot were slightly egg shaped. Which is produced by the cheap Canon 50mm (marked as Lens A). The more expensive the lens, the better quality of "bokeh balls". You can tell in the last shot that they were perfectly rounded. I guess coming from a photographer's standpoint, we really see a difference in the images because we work with such a high quality image. Expensive lenses are worth the price aesthetically speaking.
I think the video does a good job of pointing out that sharpness is not the only factor that makes a lens great and something that should be part of the test. This entire test was done at mid aperture and basically any lens is going to perform the highest at that range. But, and as the video pointed out, expensive lenses hold better image quality through f-stops as well as function better in film making when it comes to mechanics. A cine zoom will hold focus at any zoom but a DSLR zoom will soften as you zoom. At the end of the day a Toyota and a BMW are both going to get you to your location, but how you get there are different experiences.
Also, if you're shooting video at 1.8 you need the GTFO.
And yet the client/end consumer will miss every bit of it all and instead wonder about whether... (pick your production) her face was too oily; the dogs barked too loudly; the accent sounded sufficiently exotic; the hero could've jumped that far/could have jumped farther... and so on, and so on...
They should do the same test with a room full of canon and nikon fan bois and see if they can spot the difference b/w 14bit and 11bit+7 image once its been edited and color corrected
I would say the results would be no more than if you were to guess the answer for each
The end result is what matters. It doesn't matter if you used a hammer or a rock, but the proper tool will be easier to use, and make the job easier.
In my opinion these tests are kinda pointless and convey a wrong message to people who don't have a lot of knowledge or experience in this industry yet and are trying to educated themselves properly. Its all about horses for courses.
Someone who cant afford to rent a set of Zeiss UltraPrimes will use whatever lenses they can afford and work with them, do the best they can to get the most out of those lenses, work with their limitations and that's fine. Someone who CAN afford hiring the UltraPrimese, would never in their right mind choose stills glass over the Zeiss for very obvious reasons.
These tests are like comparing a ford festiva to a Maserati and arguing that both are cars with 4 wheels and will both get you from A to B even though the Maserati costs 100x more, that's not what its about with this kind of equipment.
Cinema lenses win every time in my book.
Lets also not forget the end usage here. most people shooting with stills glass will have their images/ video end up online or maybe on a magazine and that's a very small physical size area. whereas most stuff that is shot on expensive Arri/ Panavision/ Angenieux cinema glass usually ends up on televisions or even better...large format cinema screens where the image needs to hold up across a much much bigger size surface.
Handling and build quality are also an issue, cinema lenses are specifically designed for millimeter exact focus precision....Imagine Michael Bay spends $2 mio. on a shot blowing up a city block and the focus marks arent 100% correct or shift during a zoom or movement. Not to mention that these lenses are of a completely different build quality and standard....even a set of 30 year old Zeiss super speeds still sells for 20K+ these days. a decent cinema lens is an investment that can hold its value for decades. A stills cinema lens is worth half its value the minute you carry it out of the store and unpack it at home.
For the layman there might not be much difference when watched on a puny little youtube screen. but when you sit in a grading room or a theater, edge to edge sharpness, chromatic aberration, contrast and unified color rendition is what its all about, amongst many other things. If I could afford it, give me an Angenieux or Cooke over any stills glass (leica maybe excepted) any day.
2 DP's watched it on a TV "not youtube" and got a combined 15/36. Thats interesting.
Another reason why cinema lenses are better for movie making?
T stops, no matter what lens you use, same amount of light will reach your sensor/film if you use the same T stop. There will be no visible variation on the lightness/darkness of your scene when changing lenses.
Also what Dr. much said, most video/film specific lenses are parfocal (no shift in focus while zooming in or out) which is a more complicated design than almost all still lenses, which are varifocal (focus changes when you zoom) this are easier to design and with autofocus not many people notice it.
Hahaha! Someone needs to do a series on focus charts!
Ok, what if instead of cheapo 50mm lenses, like the Canon and Nikon 1.8 versions, they would use the EF 50mm 1.2L and the Nikkor 1.4G. Would there be much of a difference from the Zeiss 15.000€ ones?! As for the focusing, old russian lenses, like the Jupiter and so on, really are soft to focus with...how about those?
I shoot on an FS700 and I just switched from a $2K+ Nikon 70-200 f/2.8 to a $40 used 70-300mm variable aperature kit lens. I notice zero difference and I have 100mm more focal length to play with. Best decision I've made with that camera yet.
That's interesting. Sadly I do really enjoy shooting video at 2.8 though :)
Don't get me wrong, I do too! But I'm shooting mostly lacrosse close up at 240fps. The metabones speed booster makes that lens with all its elements look real soft wide open and it's distracting at 10% speed. Primes fix the problem so I've been shooting with an old Nikkor 200mm Ai-s lens, but when I considered how rarely I actually shoot wide open on the field and how much I could use the extra length I tried this cheap experiment and it was worth it. 20,000 views later with the USA girls in Scotland no one noticed any difference. Neither did I.
I still carry my D800 with my 24-70mm everywhere though so when I want to shoot at f/2.8 I always use that camera.