Can filters ruin the quality of an image?

[Rebel]

I just bought the 70-200L f/4. I added a polarizer just the other day. I shot some cars this afternoon (first shoot with the polarizer on) and the picture quality seems about as good as my cheapest lens. Can this mean I bought a bad filter? Its a COKIN circular polarizer 67mm

[Belmondo]

You'll lose some lens speed with a CP filter, and you might kill some of the highlights. Whether or not it is the problem with your photos is hard to say. I use a CP only when I need a CP. Otherwise, I use UV or daylight filter, and only for lens protection.

[nucki]

I just bought the 70-200L f/4. I added a polarizer just the other day. I shot some cars this afternoon (first shoot with the polarizer on) and the picture quality seems about as good as my cheapest lens. Can this mean I bought a bad filter? Its a COKIN circular polarizer 67mm

well, I heard that the cokin polarizers are'nt good. but thats only what I heard. Have you tried a normal "screw on" CP?

Peter

[Rebel]

Yeah, its a screw in type. I only use it to remove the glare of the windscreen and reduce reflections in the car's paintwork. I was shooting at about 1/40 - 1/60 handheld with the ISO right down to 100 on a very diffused overcast day. I seemed to have lost all detail. I would have liked to post an image, but I was so upset that I deleted them all first.

[IncompletePete]

I use a Jessops own screw in CP filter, its pretty good quality, got some nice results from it!

[John_T]

I would say " I was shooting at about 1/40 - 1/60 handheld with the ISO right down to 100 on a very diffused overcast day." would have more to do with it than anything else. :wink:

[Muffin Princess]

Shooting handheld at aht speed, yeah, I would say you'd lose some details, you'd probably need a faster shutter to really get some good details.

I mainly use a CP for bringing out the blues in the sky and such like, makes it look a lot deeper and warmer! :)

[timmyquest]

I would say " I was shooting at about 1/40 - 1/60 handheld with the ISO right down to 100 on a very diffused overcast day." would have more to do with it than anything else. :wink:

What he said, thats much too slow for that lens.

[Tapeman]

I have circular polarizers that cost more than many lenses. Good 77mm filters are not cheap. Hopefully you will have multiple lenses that you can use them on.

[Rebel]

Thanks for all your advice, I really do appreciate it. I think you maybe right about the speed. I always thought that overcast conditions were ideal for shooting cars. Am I terribly mistaken?

[Aylwin]

No, I don't believe you are mistaken. It's the CP that's the problem. It requires you to use the slower speed. Besides, on an overcast day I don't think you need it much. Simply removing it would give you maybe 2 stops. Then there's ISO100. You could easily use ISO200-400 to allow for another stop or two. That should bring your shutter speed up to around 1/180.

[John_T]

These were made outdoors on a bright hazy day at around 1/100s @ f2.8 and no filters. I worked more with angles that gave reflections I wanted and avoided blow outs.

http://www.fredmiranda.com/hosting/data//3694/20945Ferrari.jpg

http://www.fredmiranda.com/hosting/data//3694/20945Maserati.jpg

[Cadwell]

OK, there are probably two factors at work here;

1. (as has been mentioned) you probably needed to use a higher shutter speed unless you have very steady hands. Remember the old rule that the shutter speed for hand holding should be (as a minimum) the inverse of the focal length and don't forget the crop/multiplier factor. So for 70mm you need @ 1/100th or faster and for 200mm you need 1/320th or faster.

2. Quality of the polariser. There is absolutley no point in putting a cheap polariser (or indeed any other kind of filter) on a good lens. If you do that, you may as well have just bought cheap glass in the first place.

As to the other points, I don't think that overcast is necessarily better for car photography. Even lighting is the key.

Polarisers are wonderful devices for car photography. Used properly they can turn the glass transparent and minimise unwanted reflections allowing you to see the detail of the cars. Stick one on the front of the lens and then twiddle away until you get a result you are pleased with through the viewfinder. KennyG did a great example post on this subject a few months ago.

For some examples of my own http://2004.ukmotorsportpics.com/c215223_1.html all of those static car shots on the first page were taken in awkward early morning light using an EF 17-40mm f/4.0L and a Hoya 77mm Super HMC-Pro1 Circular Polariser. Whilst by no means perfect, I wouldn't have got them without the polariser.

[toddb]

OK, there are probably two factors at work here;

1. (as has been mentioned) you probably needed to use a higher shutter speed unless you have very steady hands. Remember the old rule that the shutter speed for hand holding should be (as a minimum) the inverse of the focal length and don't forget the crop/multiplier factor. So for 70mm you need @ 1/100th or faster and for 200mm you need 1/320th or faster.


Oh man, I thought I just had this figured out. I thought the crop factor only meant to show what equivalent frame filled picture and that the image is actually the same, except since the sensor is smaller then 35mm film that only a "cropped" part of the image coming through is recorded. In another thread I thought it was explained to me this way and therefore the DOF calculations are not with the added crop factor but of what the lens actually is. Wouldn't this be true for the inverse focal length rule as well? I mean it's the same image and distance coming through the lens as it would on a 35mm format, just cropped. I would like to know, I'm not saying your wrong since I don't know for sure.

[Cadwell]

Oh man, I thought I just had this figured out. I thought the crop factor only meant to show what equivalent frame filled picture and that the image is actually the same, except since the sensor is smaller then 35mm film that only a "cropped" part of the image coming through is recorded. In another thread I thought it was explained to me this way and therefore the DOF calculations are not with the added crop factor but of what the lens actually is. Wouldn't this be true for the inverse focal length rule as well? I mean it's the same image and distance coming through the lens as it would on a 35mm format, just cropped. I would like to know, I'm not saying your wrong since I don't know for sure.

No, I think you still have it figured right. The inverse focal length rule really relates to the Field of View which of course narrows with longer focal lengths. The narrower the Field of View, the more apparent vibration from hand holding becomes. The crop factor has the effect of narrowing the field of view so it needs to be taken into account in the calculation.

[Mitch]

Toddb,
We need Bob Gross to comment on this one. I, too, am in a quandary about increasing the shutter speed by the crop factor. My understanding of the 1.6 crop factor geometry is identicle to yours. But, the more I think of it, the increase may be correct. If you take two shots of the same subject from the same distance with the same focal length with the same shutter speed (equal to the 1/focal length) but with one camera with a full sized sensor and the other with a cam with the 1.6 crop factor and you then print a standard 4x6 (assuming the same amount of hand shake), what you have is an enlargement of the center of the image from the cam that had 1.6 factor. I've got to conclude that the hand shake would be more evident - its magnified. Right? If so, that may well support the shutter speed increase due to the crop factor - but is the increase necessarily "equal" to the crop factor? Not sure on that one either.

[toddb]

I think that is why I am confused. Would the shake be just as bad then if you took that same focal length on a 35mm and just cropped it to the 1.6 crop factor. So in a sense, that middle part of the 35mm film would not have a fast enough shutter speed according to the inverse rule. That's the part I can't seem to get out of my thoughts on this. The way I see it, and I may be wrong, that the image on say a 50mm lens is still showing 50mm perspective...just that part of that image is spilling outside of the sensor and the 10D is only capturing that center of the image. My logic tells me that if this where the case (and maybe what I just said is where my logic fails) that the inverse rule applies only to the lens's focal length and does not matter what the crop factor of the sensor in the camera would be.

[Roger_Cavanagh]

The issue with minimum shutter speeds for handholding is based on the assumption that two pictures from cameras with different crop factors are both enlarged TO THE SAME SIZE. Since the smaller sensor (10D) must be enlarged more than the larger (say 1Ds) any blur caused by camera shake becomes more noticeable.

I have looked and cannot find an original explanation for the inverse focal length rule for 35mm photography, but I reason it is based on the same logic that gives us different Circles of Confusion for different degrees or enlargement and a 10x8 print is assumed. In fact, most of scrutinise our digital images at much greater enlargements: a 100% zoom on a 21" monitor equates to almost a 40x increase.

No question that using 1/40 with a 70-200 lens is too slow, although the inverse focal length rule isn't a rule, but a guideline and some poeple will be able to manage to get sharp pictures with slower speeds, while others will need a more conservative approach.

Regards,

[toddb]

So, say I know my 135mm lens just isn't enough reach...I know I'm going to crop it in post process....say my crop will then be something more like 5x instead of just 1.6x. So then if I'm using this guideline, I'll do the math of 135x5=675. So I better make sure my shutter speed is 1/675 or better to get safe results.

:?:

So I guess it's how much the original can be downsized to fit the output size which will make it more crisp looking because it's more condensed....compared to stretching the pixels out further to make the output size. I guess I can start to see it this way.

I'm so interested in this because I know allot of my out side shots tend to be a little soft...I know it has allot to do with how far away my subject is and I'm trying to figure out how to get the most out of the camera.

[Rebel]

Wow, this has certainly caught fire. I'm begining to feel like I'm the one who's going off topic here by asking, how do I know if I have a bad CP? Its a COKIN 67mm made in France. I took some snaps yesterday at the same location of moving cars and I did get better results without the CP. But once again, this was just before sunset in a shady area.

Anyone know anything about the COKIN Filter? One guy here says he herad that they were not too good. Anyone else? This is my first L lens and I'd hate to bugger it up with a crappy filter.

On the point of the shutter speed, if I'm shooting at optimum speed, I get bugger all motion blur. Everything just freezes, especially the wheels. Because I'm shooting on public roads (a very quiet large roundabout) the cars can't really go much faster than 40 kp/h[/img]

[toddb]

I'm sorry, I didn't mean to drive this off topic.

I have a hoya CP and I like it quite a bit. I did however got it stuck on my 28-135IS....so I thought. I was trying to use rubber goves, anything to get it off. Turns out if I just turn the CP (the part that turns) to the right posistion, it unscrews very easily. Wierd eh?

Another note that I'm not possitive, but I think I remember reading that the CP makes the most difference when the sun is not at high noon....so like at a 45 degree angle. I have notices that turning the CP at high noon I can't see much difference in the sky, but say around 3 O'Clock or something, it's more noticable.

[robertwgross]

1. Make sure that your polarizer is a circular polarizer and not a linear polarizer.

2. Make sure that you know what to expect out of the exposure metering in your camera. For any given scene and it's normal metering, if you add a circular polarizer on, then the meter should shift by up to two stops. Depending on which exposure mode your camera is in, that could leave you in a bad place for shutter speed.

3. Circular polarizers get their maximum effect when the camera is pointed 90 degrees off from the sun direction. If I have a circular polarizer on at noon, and then I notice that I turn the ring and get no change in effect, then often I will remove the circular polarizer altogether so that I don't pay the two-stop penalty.

---Bob Gross---

[John_T]

Rebel, it would have helped a lot if you had told us from the beginning you were shooting moving cars, whereby you wanted to show car detail while still showing some motion. If this is the case, other than the f stops the filter costs you, the filter is secondary. Shooting moving cars handheld at low shutter speeds is a triple whammy.

Further, a CirPol needs some constants in the direction of light, as stated before, 90 degrees to the light source to be effective. On a cloudy or hazy day, the light is diffused in all directions making a CirPol practically useless, only taking away light that you need for higher shutter speeds and f stops..

Would you like to give us a more complete idea of what you are trying to do so that we can be more helpful?

[robertwgross]

Thanks for all your advice, I really do appreciate it. I think you maybe right about the speed. I always thought that overcast conditions were ideal for shooting cars. Am I terribly mistaken?

Overcast conditions mean that there are no bright sun and dark shadows (and that might be good), but if you don't have enough light to get a proper exposure without dropping the shutter low, then why in hell aren't you using a good stiff tripod?

---Bob Gross---