jsenear Loc: Hopkins, MN.

Over the years I have noticed a difference in identical lens's from the same manufacturer. Example: My son's Nikon Kit lens 18-55 VR is not nearly as sharp as mine. I"m talking about testing the lens on the same camera. Also both lens's were brand new when the test was done. We even did the test on two 50mm Nikon Prime lens's 1.8 and in this case my son's was much sharper. The tests were done with my D-50 and my son's D-40. The results were the same on both cameras. I realize the tests were subjective but I think there is some validity to this noticed variance.

jsenear Loc: Hopkins, MN.

Forgot to say tests were made with the same camera settings.

BigDaveMT Loc: Plentywood, MT

Until the factories are totally automated with no human interaction, you're always going to have some variations. Maybe one lens was made on the day after a big holiday and the folks on the assembly line were still recovering from the party. Who knows, the human factor introduces many opportunities for quality variations.

jsenear Loc: Hopkins, MN.

Kinda makes our recommendations for lens's moot.

BigDaveMT Loc: Plentywood, MT

Some of the variations are going to be more noticeable than others. Just look for those lenses which have significantly more positive feedback than negative.

glojo Loc: South Devon, England

IMPORTANT...THESE ARE QUESTIONS AND DEFINITELY NOT OBSERVATIONS\REMARKS or CRITICISMS

This post has really got me thinking about how on earth you can conduct a test that would be so exact.

1. Are you using RAW files?

2. Is it in a controlled lighting environment as surely the very slightest change in conditions would cause a difference?

3. Are you downloading the RAW files onto one computer?

4. What about the computer cache, is that going to remember any settings?

5. I take it you used both lens on the one camera and then conducted the same test on the next camera?

6. Do you use the same computer software and same graphics card, same type of memory as RAW files are quite large?

7. Were these tests just visual or did you use software that was capable of detecting the differences?


These are just some of my thoughts being put to paper as I am guessing it is possible there are differences but are these things assembled by robots who do not have an 'off day'?

jsenear Loc: Hopkins, MN.

I knew those question would be asked. Tests were done on the same computer and same lighting conditions. My son's back yard. The only delay taken was switching lens's. What does that take. 30 seconds? This test was done with Jpegs. Results might be different with raw files but I don't think so. Computer was not shut down and no other work was done between tests. Cache should be the same.The computer might be a variable but these tests were not computer dependent. The memory card (SD) was put into my son's Selphy printer and prints made right off the SD card. Results were the same. The tests were visual. That's why in my post, I said the tests and results were somewhat subjective.

jsenear Loc: Hopkins, MN.

I know the slightest light change would make a difference. The photos were taken one after the other. No clouds or anything else to deflect the sun's rays happened. No new shadows etc.etc. I think there sometimes is a large difference in two identical lens's. Even from major companies like Nikon and Canon.

Iowegan

Wife's camera and mine are identical models, D7000.
We each have identical Sigma 18-200 lenses for general usage.

All other variable's being equal, using single point focusing, her camera would repeatedly present better focused images. :)
And, it was actually her camera body that was the deciding factor here. Switching lenses produced identical results.
Her's were sharper images, across the range of 18-200mm.

I did the AF Fine Tune shortly after, and found I had to compensate, whereas her's was spot on at 0.

You never know what you're going to get here. Sometimes, it's not the lenses causing issue, but, the camera.

jsenear Loc: Hopkins, MN.

Problem is we use the same camera for the tests. First the D-50 and then the D-40.

Iowegan

It may not necessarily be the lens, then.

There was thread awhile back, where the discussion was based on similar thoughts. Focus issues with lenses and cameras.
A member stated he'd sent a lens back to the manufacturer, perhaps multiple times, I don't recall, exactly.
If I remember correctly, the topic of AF fine tune came up, and was properly and thoroughly thrashed about. :)

Point is, perhaps, it's not the lens. And then again, maybe it is.

Perhaps someone else you know with a similar body to test with, to see if there are similar results, to point one direction, or, the other?

glojo Loc: South Devon, England

Hi jsenear,
Thank you for taking my questions in the spirit they were meant.

I have no idea what size the files are for your cameras but if we assume the 50D takes a 15mb RAW file and the 40D takes a 10mb RAW file and you have elected to download in jpg format then instantly these files get compressed. This compression is removing information and it does this in a random manner. The randomness will clearly show discrepancies because of the very nature of the beast. NOTHING to do with the lens at all. ;) (observations and NOT point scoring)

You are not being subjective in your observations of what the camera sees. You are being subjective on the subjectiveness of how the camera opts to compress each single file!! :shock: In other words you have quite correctly noticed the differences, but this is probably due to the use of jpg files.

You may or may not have expected this type of reply but the reality is you are looking at files that have been shrunk from possibly 15mb to at the very most 4mb?? What size in mb are those jpg files you are comparing?

Again I am NOT being critical, I BELIEVE you are seeing variations... I am simply curious as to how you are reaching decisions that are blaming the lens when to me it is probably the compression that is the cause?

penguinpete

Another point to consider is the concept of tolerances. Even if factories were totally automated, there would be a problem with tolerances. Engineering specifications always include 1 or more tolerance specifications, usually expressed in fractional units plus or minus from the specified dimension. The accuracy of a lens, or of a camera body, is going to depend on how all of the tolerances combine. In some cases the tolerances will offset each other and result in a truly sweet performer, and in others they will accumulate in one direction or another and the end result is a real dog. If the quality control "window" of acceptable performance is too wide, there will be noticeable variations in performance between "identical" lenses or "identical" bodies, but the manufacturing costs are lower. If the "window" is too narrow, rejects and cost go up and variations from one unit to another are less pronounced.

jsenear Loc: Hopkins, MN.

Thanks Penquin Pete. I think that's the case with our tests. The bad 18-55 was gotten rid of. Both 50mm's were kept and reside in our kits. One is better than the other however. Just wanted to throw this out to the list when we are talking about lens's and lens comparisons.

Screamin Scott Loc: Marshfield Wi, Baltimore Md, now Dallas Ga

Sample variation has been with us forever....It has been noted & discussed many times over the years. We used to comment about it back in manual focus lens days as well..