Depth of Field Matters (1 Viewer)

When using the f7.5 ED80/600 Skywatcher on my DSLR, the depth of field seems quite short compared to that achieved by commercial TV broadcasters with their telephoto lenses when televising outside sporting events (cricket, rugby, tennis, etc).

Has anybody else noticed this, or do I need some new glasses?

I am referring to the ground based cameras and not the flying skycams that zoom around the playing area on overhead wires. Even using the endcap on the dew shield to decrease the effective lens diameter for an approx f11 on the Skywatcher does not seem to come anywhere near that achieved with OB TV cameras.

I understand for a given crop factor the Skywatcher depth of field is the same as would be achieved if I used a similarly dimensioned, very expensive, lens from Canon or Nikon.

I have read that crop factor effectively increases a lens' s magnification without a resulting a reduction in the depth of field.

Do commercial TV OB cameras only employ a 1920x1080p, physically small, very expensive, sensor with a large crop factor to achieve such large magnification ratios with what I regard as a remarkable depth of field compared to my DSLR and Skywatcher?

The TV picture still shows the large image distortion (foreshortening) associated with high magnification ratios. The low light performance of OB TV cameras seems to be very good so their light gathering ability is very impressive. As you would expect, there is no sign of visible noise at any workable light level on the TV camera.

I realise that comparing the Skywatcher with professional quality TV hardware is not a "like for like" comparison given the enormous price differential. The Skywatcher is very good quality for an affordable price.

Commercial TV cameramen are VERY good at focus tracking and very rarely have out of focus issues. I have heard stories that they spend countless hours practising focusing on a swinging bottle at the end of a length of string before they are let lose in a production environment. Having a good depth of field also helps focus accuracy and the range of focusing tools available in a top end, full professional quality, very expensive TV camera far exceeds what is available on most DSLR cameras.

If my observations are correct, how do TV OB cameras achieve such a good depth of field compared to what we achieve with the Skywatcher? What optical principles are at work to enable the TV camera to achieve its depth of field? What size are the lenses employed on OB cameras? How many pixels are there in an OB camera sensor?

TIA,
Brian

I don't have any technical lens expertise so I cannot comment on your post but I also find that the SW80ED has extremely shallow DOF and it makes focusing very difficult. It doesn't forgive; any small error in focusing will give a OOF photo. That's why I don't rely on my eyes to focus anymore; I always use focus peaking.

Brian,

Yes the DOF at these focal lengths is shallow.

To assess the DOF you may find the "DOF master" tool useful.
http://www.dofmaster.com/dofjs.html


To give an idea, below are some values returned by the tool. Say you use a Nikon/Canon APS camera, for which recommended circle of confusion is 0.025mm. I think the endcap on the Skywatcher reduces with 2 F-stops so more like F/15, I haven't measured precisely.

I created a table with arbitrary (useful) distances, from long range/50m down to 6m which is about point blank when fitted with 80mm extension tube.

Distance - DOF @F/8 - DOF @F/11 - DOF @F/16
50m - 2.7m - 4m - 5.5m
25m - 0.7m - 0.9m - 1.4m
12m - 0.15m - 0.2m - 0.30m
10m - 0.1m - 0.15m - 0.2m
8m - 6cm - 9cm - 12cm
6m - 3cm - 5cm - 6cm

F/8 is the closest value supported by the tool. Since the DOF is linear to F/number the DOF at F/7.5 is about 6% shorter compared to F/8.

Anyway, this gives an idea of how shallow the DOF is when shooting at close range.

Whenever possible, I use the 5X-14X magnifier on my camera (E-M5) to support me in adjusting focus precisely. As Jules writes, unless you pay attention, ending up with OOF pictures is easy.

BTW, do you have a single or dual speed focuser? And what camera are you using?

HTH
Tord

DOF Matters

Tord,
Thank you for the infomation re the DOF tool. I had not seen this one before.

I have a Canon 600D with a single speed focuser on the Skywatcher. I also use the x10 magnifier in Live View mode to make up for limitations with my eyesight. However, it is not much use on a bright day when you cannot get the camera into the shade. I understand the 600D to have the Canon APS-C sensor with a crop factor of about 1.6. I use arpeture preferred mode, together with a cable release on the 600D to reduce vibration.

I believe your E-M5 to have a 4/3 sensor with crop factor of about 2. The way I understand things, your E-M5 would give a greater magnification than my 600D if we both focused on the same bird in the same location with our SWs. Both cameras would have the same DOF using the Skywatcher.

I think the Skywatcher is very good value for money and I can live with the shallow DOF in view of its ability to deliver great quality images at a very reasonable price. It would be nice if I could increase my keeper rate and I understand we all have the same problem.

What led me to start this post is that I am impressed with the DOF the local TV station is getting with the OB cameras covering the cricket match with India here in Brisbane. The only way I think they can be getting such good results at high magnification is with a large crop factor and a very small sensor. I am interested to learn if other optical mechanisms are involved. I don't know what geometry the TV camera sensor is and was just looking to see if anyone here could supply some values out of interest. The TV coverage seems to have much better DOF than I have been able to achieve with the SWED80 but I may not be correct.

In HD mode, our TV can handle 1920x1080p. When I shoot this format video with the 600D the results are rather nice. The 600D's 5184x3456 sensor is 3:2 aspect ratio and it would be interesting to know how Canon use these pixels when producing the 1920x1080p 16:9 format images. Do they combine pixels or just throw some away? Given the TV station's budget I suspect they would have a very nice sensor in their OB cameras. There is no way I could match their budget but I am interested in the optical principles involved as the TV OB cameras seem to have a much better DOF than I have been able to achieve with the SWED80. Do the OB camera sensors shoot in a larger format frame than 1920x1080? Any assistance greatly appreciated. I am a beginner when it comes to long lens photography but would like to learn more.

Thanks again,
Brian

Brian O'C said:

Tord,
Thank you for the infomation re the DOF tool. I had not seen this one before.

I have a Canon 600D with a single speed focuser on the Skywatcher. I also use the x10 magnifier in Live View mode to make up for limitations with my eyesight. However, it is not much use on a bright day when you cannot get the camera into the shade. I understand the 600D to have the Canon APS-C sensor with a crop factor of about 1.6. I use arpeture preferred mode, together with a cable release on the 600D to reduce vibration.

I believe your E-M5 to have a 4/3 sensor with crop factor of about 2. The way I understand things, your E-M5 would give a greater magnification than my 600D if we both focused on the same bird in the same location with our SWs. Both cameras would have the same DOF using the Skywatcher.

...

Thanks again,
Brian

Click to expand...

Brian,

The DOF is linear to the Circle of Confusion (CoC), as well as to the inverse of the diameter of the aperture. 4/3 sensors have a "crop factor" of exactly 2, whereas APS sensors are somewhere around 1.5... 1.6. So answer to your question is yes, the field of view for a given piece of optics is narrower (i.e. larger magnification) on a 4/3 camera compared to APS.
Recommended COC for 4/3 is 0.015 mm. This value may be outdated though, since the original 4/3 sensors had 10M pixels, today's 4/3 sensors have more like 16M pixels (26% extra pixels in each direction) in which case the COC is probably more like 0.12. I may be not be totally right, though.

Tord,

I am starting to think the TV OB camera achieves its apparent wide DOF because it uses digital zoom. There seems to be a lot of medium priced video cameras using the 4k format, which I understand to be 3840x2160. If you had an even larger sensor, I suppose you could employ greater amounts of digital zoom while still using the 1920x1080 HD output format without having to fudge pixels.

My 600D with a 5184x3456 sensor can apparently support a digital zoom of 2.7x when shooting 1920x1080p video @ 25fps (see http://www.cameralabs.com/reviews/Canon_EOS_600D_Rebel_T3i/movie_mode.shtml). When printing a 12x8 I normally scale my still images to 300dpi or 3600x2400. However, at home I never display my bird photos on a monitor over 1920x1080 (I don't have anything better, and my notebook which I use for PP is only 1366x768) so maybe I am not using the 600D to its full capability when using a screen based display for stills.

From what I have been able to find out, it appears the 600D has a video snapshot mode that lets you take a 2, 4 or 8 second video burst. It seems as though multiple snapshots can be concatenated into a single "album" file. I have not tried extracting a single frame from a video clip but there seems to be lots of software that support this operation. The snapshot mode appeals to me because I can take a burst of 50 frames at 1920x1080 with 2.7x digital zoom and extract the individual frames and then perhaps try some focus stacking.

If I use the digital zoom, what happens to the circle of confusion which I think may have something to do with DOF? Do you think DOF would vary with digital zoom, please? I don't have a working knowledge of COC. The physical pixel size is obviously not changing.

The 600D has a quoted digital zoom range of 3-10. Anything above 2.7x would require pixel creation and hence I think a resulting reduction in quality. The digital zoom on the 600D is only available in 1920x1080 video resolution. For screen based display of bird photos, where I don't believe we are using the full resolution of our SLRs (unless you have a VERY good monitor), do you think there would be much noticeable difference?

With Xmas approaching and our daily maximum temps running well above 30C I don't think I will be able to do any meaningful tests for a while so I am just bouncing ideas around at the moment. I realise most people will scream blue murder at the thought of using digital zoom but as long as pixel fudging is not required and the resulting image is as big as our display device, I wonder if it does all that much harm - especially if it gains us some DOF improvement. Unfortunately my camera does not support digital zoom for stills. Would be interested to hear any comments you may have or experience you can relate.

TIA,
Brian