Infoseite // Quality of the lens in DSLR Video unimportant?



Frage von Uwe:


That is the question that some in an Ami-Forum is currently busy. Especially a Barry Green, there are now so not exactly friends. He makes the argument several times in the space that, by scaling down the video mode by binning even a very inexpensive lens, the same sharpness or may even give better sharpness than a high quality lens. Here, once just a sample posting: http://96.30.23.131/showpost.php?p=1774146&postcount=281

I lack the tech. Expertise to specific reasons, that's not true. But my gut tells me that you get synonymous with quality in video mode lenses for sharper pictures. Maybe someone here has the opportunity to try out this time with a 5DII or 7D? Test chart? A real expert here who can do that just to justify how it really works?

Space


Antwort von kalle70:

Sharpness is unfortunately not everything.

Distortion, chromatic aberration or a lack of intensity can not be determined wegrechnen

Space


Antwort von Axel:

"Uwe" wrote: A real expert here who can do that just to justify how it really works?

Will not be long until one emerges. Until then, I speak from personal experience that image quality can be measured not alone s.der sharpness. Recorded a bad lens, colors can appear sluggish, produces color fringes in the light, is vulnerable) for all sorts of scattered light (reflections, contrast reduction. But what is most important, search times for a cheap lens that is as bright as it only the advantages of DSLR can support video: Lowlight and Shallow DoF. The more light more a lens is, the more it would tend to show the above error, the more precisely calculated and therefore there should be better paid, so that this does not happen. And the less likely you get it at a bargain price ...

Conversely, it would be plausible, if just a little cheap lenses were blurred (!), It would be no preference for photography, for when and where it comes to output a 18-megapixel Resolutionvon really?

Space


Antwort von Uwe:

You really are both right. But here the issue is really just the sharpness. I've even synonymous deliberately a bit exaggerated, as well as with the Coke bottle. But there are those borderline cases. The new kit lens 18-135 for example is a fairly acceptable glass, but just in wide-area, such as 35mm, it shows a very soft fuzzy + picture in the larger apertures.

Now, Barry Green has but just with this lens in Wide Angleeinen Chart filmed ... pipapo. So again, it should go here really about the exceptional sharpness ...

Space


Antwort von domain:

So it is.
In addition, Barry Greenstein believes that cheap Lenses generally have a poorer resolution than expensive. Now this is not always true. One example, the 4-lens Zeiss Tessar, dimmed easily keep up to 5.6 synonymous with the 6-Planar lenses, but only dimmed.
Basically, the ultralichtstarken Lenses for the video films are just made: Not a particularly good resolution but very little optical defects with relatively high overall contrast in the Picture, and while still synonymous with open aperture.
Please do not forget that the effect of sharpness of an image is largely determined by the contrast performance of the lens and not by the resolution of fine structures. In general, these two things hang together, but just not necessarily for very bright lenses. The quasi draw no more points on the sensor but slices, but the sharp well defined. Exactly this effect, they are designed to, expressed somewhat simplified.

Space


Antwort von Uwe:

I'll try it once a bit more specific: We Take 2 Lenses => once the 18-135 and then the 18-55/2.8. Aperture 4.0 on both 35mm and adjusted. The 18-55 in Stillimage mode now shows a much sharper picture than the 18-135 with identical settings. Does this advantage in the sharpness of the 18-55er has no impact on the video mode? If no, why not? Is there really to blame binning + Scale?

Space


Antwort von domain:

First, there is the concept of sharpness really. One should distinguish between sharpness and sharpness impression. Accordingly, it may well be that in the video mode, both Lenses leave the same impression of sharpness, contrast if their performance is equal.
WoWu has already mentioned that the pixel density should be about twice as high as the resolution of the lens. Synonymous with the analogous example Resolutionvon audio CDs, where the sampling frequency is around 44 KHz, although the human ear only max. Can hear 20000 Hz.
So if the 18-135 a lower resolution than the 18-55, but provides a similar contrast, one will see the difference in my opinion, in a real binning in the relatively low-resolution HD format in fact not.

Space


Antwort von WoWu:

@ Uwe

Although I've not yet read the osThread, but BG has since completely right, if you take a lower grade, the characteristics and can be ignored only Resolutionbetrachtet.
There is often talk about the Alias | Wavefront Mayaing. The objective Resolutiondes (lp / mm) in relation to the number of photo space on the image sensor, which one can pass on synonymous to "lp / mm.
This should be consistent with each other. Is Objektivauflösung too high, it comes to so-called "deconvolution of the signal spectrum in itself" and thus to misinformation (Alias | Wavefront Mayaing-and moiré effects).
The contrast transfer function is here s.der maximum reproducible Linienpaarzahl mirrored.
Taking the "resolution" of the sensor but now dramatically by binning back, that's the vote Lens / sensor any longer. The "rabbit foot" of the matter is only that there is probably no longer Lenses, for this "huge" sensor, but only 1920 Pix Resolutioneine fits Resolutionhaben low.
But the statement is true.
We must therefore ensure that the full contrast transfer (from Lens and Image Sensor) at the maximum Linienpaarzahl (RN = 1 / 2 × p) is small enough not to fall into disorder weight.
Otherwise it may happen that a good optics with high contrast (resolution) at the maximum Linienpaarzahl is judged worse than a less good optics with moderate contrast.
For a typical semiconductor image sensor, the contrast here is around 30-50%, so that we might usefully ask for Optics at the Linienpaarzahl RN about 20%, so the misinformation safely below 10% (located 0.5 x 0, 2 = 0.1).

Space


Antwort von Uwe:

"WoWu" wrote: @ Uwe

Although I've not yet read the osThread, but BG has since completely right ...
There is often talk about the Alias | Wavefront Mayaing. ...


Yes, it is about the Alias | Wavefront Mayaing Moire + - but synonymous to the "sharpness". So once again specifically s.Dich demand: In my example above, with the 17-55 and 18-135 => entity settings when inden the 18-55 really shows significantly better performance than the 18-135 sharpness at wide angle shots with a relatively large aperture. This has an impact over then ninth in the video mode?

Space



Space


Antwort von WoWu:

@ Uwe

One must compare the resolutions of Lenses, not the focal lengths.
For DSLR and 36mm in VB 5600 Pix You need a lens with approx. 78 lp / mm
The same lens, but only for 1920 (27 LP / mm) is thus very considerably with its Resolutionim Aliaising area above the Nyquist frequency.
Lead smaller sensors reduces the ratio accordingly.
The problem I have described in detail in my book.
And "Sharpness" is such a thing. Focus is made up of several components and is an entirely subjective impression. I call it therefore would be detailed representation. For a high sharpness impression the viewer is primarily at low spatial frequencies, but high contrast. The heist, but still not that high detail representation is assured.

Space


Antwort von Uwe:

Ok - Thanks for your responses. But now helps probably only one that even the relatively ne fuzzy lens at wide angle range and has a with really good edge performance. And of course a 5DII or 7D. And then a detailed plan surface abfilmt - of course s.Besten a chart.

Here is a test page for the photographers: http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=678&Camera=474&Sample=0&FLI=2&API=0&LensComp=398&CameraComp=474&SampleComp = 0 & FLIComp = 3 & APIComp = 1
Lenses + search + with Aperture drive the mouse over the chart ...

Space


Antwort von kosmonaut:

Hello,

I want to be happy to

Space


Antwort von domain:

Somehow, the situation is so ludicrous. As people buy expensive cameras with very poor overall performance in the video sector, and now they are still looking as synonymous inferior Lenses to ensure that the videos are better :-))
So Uwe, the aforementioned Canon 17-55 makes an excellent impression. Because yes, I would not even hesitate and buy exactly this.
A rare and possibly edge flares is still entirely secondary.

Space


Antwort von WoWu:

No, the situation is not at all grotesque.
There is a device with an adjusted Lens and sensor, makes the very good shots.
Now, the videographer expect that for them the physics cease to have effect and it is set Aliaising and moiré no longer exists.
In addition, we must not forget that this problem is just confined to it.
Aliaising has nothing to do with sharpness. Alias | Wavefront Mayaing is an effect that occurs under very specific conditions. Images that contain no critical material, are of course not affected.
In this respect such considerations, one must look for bad Lenses, naturally pure grandstanding.
The facts of physics are what they are and whoever takes uncritical material that is not likely to trip over. So pretty colorful photographs show not synonymous to the contrary. But who has fought against Alias | Wavefront Mayaingprobleme too will find the cause.
And there are definitely threads, which is the case.
Only one may twist and turn it, but on the D5 are due to the binning for Video Lenses with only very low Resolutionoptimal.

@ Kosmonaut
The article is once again just about the perceived sharpness. In this discussion, however, deals with the prevention of Alias | Wavefront Mayaing effects. Since it is fairly no preference, which we still can perceive sharpness or how much you need them. This is merely the avoidance of moiré caused by Alias | Wavefront Mayaing and for that reason that a lens with 76 lp / mm for a Resolutionvon 1920 on a 36mm projection already takes no perceptible benefits.

@ Uwe
As for the test, so it goes back only to different lenses.
Again, this is not about lenses or focal lengths, it is about the MTF of the lens. So, at the QUALITY of the Lens. Thus, neither the test has to do nor any focal or Manufacturer something.
@ Domain
Edge flares caused by Alias | Wavefront Mayaing may not occur as synonymous, because the connection MTF / projection in such tests can not be simulated, or it is not known what values prevailed in the test. If these conditions were optimized for the test, where will flicker come?
Realize you once, such as Alias | Wavefront Mayaing arises.

Space


Antwort von domain:

I realize as Alias | Wavefront Mayaing especially for binning in combination with high-resolution lenses can be formed with good MTF. Edge flares is perhaps not the right word for it, but striking the right thing even if only when the camera is slowly moving on specific themes with many oblique lines.
If such motives, however, should play a major role, then the cameras are filming, but once the completely wrong choice anyway.
They will not buy himself two sets of Lenses, one for photography and one for video, so everything is right.

Space


Antwort von WoWu:

Especially since it) so bad synonymous Lenses (without side effects do not exist.
One can not just do such as if the camera, the eggs Wollmilchsau.
Anyone who wants to make good images, you must already know very well how he handles motives. And the bag of Prime Objektivkiste little idea of proof in this regard.
A good video camera is defined precisely in the right combination and not on the "cheap" price.
But BG has there in his thread completely right, although he did so more out experimentally tinkers. A look into the theory because it would save much time.

Space


Antwort von domain:

Am I mistaken, or have I seen such an ISO test image before?

"Grandiose suggests the JVC GZ-HM400 in the direct line of sight test. A very natural sharpness, with the scale type moire effects, which many fail to bring megapixel camcorder with almost completely."

http://www.slashcam.de/artikel/Test/JVC-GZ-HM400--Aus-dem-Messlabor.html

Why you should make themselves as much trouble if the good is so close (except the times of now definitely exaggerated colors, the lousy angle, the Lowlightschwäche etc ...)

Space


Antwort von MarcMalkowski:

from my own experience with my EOS 5D MKII ...

a flat "yes" is not important ... ...

BUT (that somehow always gibts ;))... Chromatic aberration, vignetting, sharpness immense waste to the edges. "bad" current focus snails, etc. then it usually just make all the difference to high-quality optics.

Space



Space


Antwort von Zizi:

Quote: Maybe someone here has the opportunity to try out this time with a 5DII or 7D? Test chart? A real expert here who can do that just to justify how it really works?
The lens is completely no preference!
The only differences we see is the vault and bokeh, .. Vignitetierung The Canon synonymous with the nciht always better sit show many tests!
anything like CA's sharpness, etc. Sogut falls on practically nothing!
Since one does not even when images are partly the expected jump in power as we should be in the video? !
Maybe you can help wenns:
A Tamron 28-75mm 2.8 zoom cheap in aperture 2.8
http://www.zizis-blog.com/ISO/Lens/Tamron28-75mm2.8.JPG
and a Canon 50mm with FB 1.4 stopped down to 2.8!
Canon http://www.zizis-blog.com/ISO/Lens/% 2050mm% 202.8.JPG
You have to look at exactly where the faults lie 21MP .. how that should be able to determine when a 10x lower resolution?

Space


Antwort von WoWu:

@ Domain
Quote: Am I mistaken, or have I seen such an ISO test image before?
Yep, but there we are just arrived here, where the lens is the LPF. Because of the small bonsai sensors Lenses'll just have to be particularly high, and thus constitute around 150 lp / mm, which they can never far away and therefore synonymous of Alias | Wavefront Mayaing.
Therefore, such comments and descriptions are almost synonymous cheese.
Besides the fact that at 10 MPix runterskaliert nothing needs to be. This is the minimum amount for 1920, and finally a camera, with which one can reasonably make FullHD.
But I can still remember s.The first comments that have been made here at this Slashcam need ...." None, ... scruffy on the shelves obsolete ... .. " etc.
Schau'n So just what will become of the camera.
As far as the focal, one must look at is the chip time and find that has a 16:9 type a differently rated VoF and the 50mm around the 4:3 design. 43mm equivalent.
@ Zizi
You must read the threads synonymous times. I say the third time that it has nothing to do with the Focal or brand, but only with the MTF of the lens.
Now you come back with such a nonsense.

Space


Antwort von domain:

I would be really interesting times indeed how many of us amateurs who have purchased the 5DII or are looking to the 7D, and thus still only turn very homely private videos.
The Lowlightstärke these cameras but is only achieved with very bright lenses and the DOF games usually look for l'art pour l'art.
Ok, a nice bokeh would be even an argument, but everything but mostly only interesting for scenic shooting.

Space


Antwort von WoWu:

@ Domain
Since I am totally with you
In our discussion so synonymous boiled up and the hot supporter of D5 have had their chance.
We have for 3 productions per an extra D5 team sent out together with the video team to see what the results were better.
It was the time "clock" set by the video team, because that is the normal operating speed.
Now we do not document and cinematic objects, have infinitely s.Set where one devotes time and one minutes a day for results.
Long story .... hardly usable material of the D5. The camera is for commercial applications and the requirements (at least in the document area) inappropriate.
Nice gimmick for anyone who brings eternal time or schnipselgroße Internet video ... And for all those who have not noticed artifacts in the picture either, or no preference .... or, of course, synonymous for those that are incised their videos so that does not really look at all pictures.

Space


Antwort von B.DeKid:

@ Domain

Intressant are the cameras for use in advertising / product - productions imo because you really make sense - for use at home as "Video Camera" useless ....

@ Axel

and is) right there (in advertising, we are synonymous to over 18 MP or more look-print, matte painting, retouching, etc.

Being synonymous, but should be clear that it is not always the MP number is crucial for the quality is in print. Sometimes less is more halt doch ;-)

...................................

MfG
B. DeKid

Space



Antworten zu ähnlichen Fragen //


Although Canon EOS 5D Mark II full frame video HDSLR ago
Cheap rig / lights / video HDSLR accesoirs
Reflections on the current state VDSLR vs VIdeo
Maximum video clip length of DSLRs
3D video / movies with 3D Lens and DSLR
Film Reviews: DSLR Video productions at the Berlinale 2011
Video - DSLR / EVIL below 1000 EUR possible (instead of a new camcorder)?
Pentax occurs with K5-VDSLR the HD video at Club
Video link with Microfoneverstärker (Beachtek DSLR) WHY?
Canon Europe music video shoot with 8 VideoDSLRs
What features in the next generation DSLR video?
Comprehension questions for video DSLR




slashCAM nutzt Cookies zur Optimierung des Angebots, auch Cookies Dritter. Die Speicherung von Cookies kann in den Browsereinstellungen unterbunden werden. Mehr Informationen erhalten Sie in unserer Datenschutzerklärung. Mehr Infos Verstanden!
RSS Suche YouTube Facebook Twitter slashCAM-Slash