Poor FoCal Result

[soldrinero]

I picked up a new 35mm IS lens in the recent Canon sale and needed to AFMA it, and while I had FoCal set up I decided to re-test my other lenses as well. I got this "poor fit quality" result from the 50mm f/1.8 STM, along with a lot of scatter at each AFMA value. Is this a normal result for this lens? If not, do I need to have it serviced, or am I using FoCal incorrectly?

Thanks for the advice!

 

[Don Haines]

Just my personal observations, but......

FoCal works best with lots of light and even lighting.... I found that I needed a pair of 500W studio lamps, one on each side of the target as a minimum to get consistent results.....

Your camera must be straight on the target..... not a bit up or down, or left or right.....

You really really really want a heavy and solid tripod.... including the head..... vibration is your enemy!

On the subject of vibration, don’t move when calibrating..... you will vibrate the floor.

I don’t know if it makes a measureable difference, but I put on the lens hood to block unwanted light....

Your target looks yellow..... are you shooting under normal incandescent bulbs? I use “daylight” bulbs.....

Hope this helps! Let us know how it turns out....

 

[takesome1]

Agree on the lighting.

A little to far from the target according to the report.

Repeatablity is key. Set it up again and compare

 

[Mt Spokane Photography]

Assuming the lighting was bright, it looks like a bad lens to me.

Set the AFMA to the recommended Value and try taking 10 Phase Detect shots at about 2.7m, setting the lens to infinity each time and let the camera do phase detect AF.

Next do exactly the same thing, except put the camera in live view and use live AF.
Check to make sure that the live view results all focused sharply, if they were all over the place, there is a lens issue.

If they are equally focused, Compare the results with the PDAF images. If live view shots are more sharply focused, you need a different AFMA value.

I'd also try the process at a closer distance, about 1.5m, the results may differ, so you will need to average the AFMA.

 

[SecureGSM]

If I may, just a couple of observations:

1. AFMA curve looks normal. The Poor result is due to far right AFMA curve point being higher than it should. Please notice a pair on cof confirmed points at +20 sitting higher then predicted curve. That’s all.

Possible cause: low level and or flickering light.
I would run the test again in good natural lighting conditions again.

2. Astigmatism level reported at around 7% is a worry. I would usuallly advise client to replace or fix such a lens under warranty. Anything over 4% is a sus. 7% is a no brainer.

 

[soldrinero]

The lighting was a mix of compact fluorescent and incandescent. I used the "Target Optimization" feature before running calibration, so it thought I had enough light, but it was probably marginal. I'll definitely try again with better/natural light.

I made sure to stand still when the measurements were being taken to avoid vibration, and the tripod is pretty decent. I'll try to compare LiveView and PDAF results, and see what that says.

Thanks for the suggestions!

 

[SecureGSM]

the higher the EV value of lighting at the time of the shot, the higher the FoCal QoF value.
fluro is highly flickery source of light, therefore some of you shots are taken at higher or lover EV level due to flicker. end result: unreliable data set. please re-run tests in a good natural light and results will be much more consistent.
there is nothing wrong with your lens.

The lighting was a mix of compact fluorescent and incandescent.

 

[Viggo]

I gave up FoCal years ago, it was excellent when the earlier versions were released and it was quick and easy and always gave me the exact correct value. Something happened and now all it does is give me frontfocus results with the same setup, and I’m veeeeery careful to set everything up right. One example is with the 1dx and 200 f2 it couldn’t get a results because it was so inconsistent that I got the error message “bad camera/lens combo” lol. Calibrated with DOT-tune and it was done in one tenth of the time and was flawless ... Poor fit and “tried five times, but inconsistent result” I got with the 24-70 II and even when I tried with 35 L II. And when the test went right, and everything looked great, massive frontfocus by 6-7 points.

 

[AlanF]

My 50mm 1.8 STM gave similar poor fits, and I am a FoCal freek. I put it down to the very shallow depth of field and not the world's best AF of this el cheapo but good lens. It's slightly soft also at f/1.8, which adds to the problem of calibration. Mine also had negative AFMAs: -8 to -12.

Some lenses are easy to calibrate with FoCal. My telephotos of f/4 - f/5.6 maximum aperture are a breeze. But at f/8 with extenders the curves are horrible.

 

[SecureGSM]

that is due to decreased contrast when shooting at a higher apertures. there is a simple trick to aid FoCal in these circumstances though: I use high power LED torch, located a little bit lower and about a meter away from the target. this will increase local contrast of FoCal target dramatically and improve your exposure by at least 3 stops at the same time. it works a treat.

TrustFire AK-91

https://www.ebay.com.au/itm/TrustFire-AK-91-15-CREE-XM-L2-LED-Flashlight-18000-Lumen-5-Modes-26650-Torches/222293849511

My telephotos of f/4 - f/5.6 maximum aperture are a breeze. But at f/8 with extenders the curves are horrible.

 

[soldrinero]

My 50mm 1.8 STM gave similar poor fits, and I am a FoCal freek. I put it down to the very shallow depth of field and not the world's best AF of this el cheapo but good lens. It's slightly soft also at f/1.8, which adds to the problem of calibration. Mine also had negative AFMAs: -8 to -12.

Some lenses are easy to calibrate with FoCal. My telephotos of f/4 - f/5.6 maximum aperture are a breeze. But at f/8 with extenders the curves are horrible.

I had an easy time calibrating my new 35 f/2 IS and my 24-70 f/4 IS. The 85 f/1.8 was a little messier, but still worked pretty well and gave a value close to what I had found previously. But the 50 f/1.8 STM was a mess - the previous FoCal result from a year ago when I bought the lens was -4, and now it "found" -15!

 

[Mt Spokane Photography]

The AF of the lens should not be inconsistent, and there should not be issues closer than 6 ft.

I'd definitely have Canon take a look while its in warranty. Since its light, it can be sent by Priority mail for the 1 lb rate.

 

[Don Haines]

Oh wow. I thought I was the only one getting -15 on my 50mm 1.8 STM using FoCal. After using FoCal, I thought about shipping it in for calibration while under warranty as my shots are fantastic about 6ft away and beyond however, closer than that, the autofocus is inconsistent.

Ages ago, I ran my 50 F1.8 through Focal and the results were terrible.....

I recently upgraded to the latest version, and after this post, decided to run it through again.... I ended up with VERY inconsistent focusing and an AFMA value of -11

This is a cheap and poorly built lens. I did not expect to see the consistency of a 70-200, but I didn't expect it to be this bad either!

 

[AlanF]

We seem to be ending up with values of ~ -10 to -15 and bad consistency. I have to say, however, at f/4 my 50mm 1.8 STM is fabulously sharp from corner to corner. I use it attached to the 5DSR for copying detailed artwork, documents etc with superb results.