MicroAdjustment Software

No problems for me with 7D, 5D2 and 1D3. When using long lenses you really must have a very stable platform. Sticking a tripod on carpet or wobbly floorboards with people moving around is not going to fly when you have high magnifications and no MLU.

When I calibrated my 1D3 and other bodies it was outdoors on a calm day (no wind) with good daylight. I used a Manfrotto 055MF3 tripod and 468mgrc2 head on a solid concrete base with the legs fully collapsed. I also added lead ankle weights draped over the body/lens for most camera/lens combinations. Where MLU was an option I set it for 2 seconds rather than the default.

I didn't always calibrate at an exact 50X focal length - often nowhere near that figure - but I did calibrate at "sensible" focus distances for each lens and in all cases stayed well clear of MFD.

You also need the lens to have consistent and repeatable behaviour. Being out of calibration is one thing, but if it keeps landing in a different spot each time you focus that is going to make the results uncertain at best. That was my experience with my 50/1.4. The AF simply is not precise enough to give repeatable results shot after shot after shot. Nonetheless you can follow an approximate trend line and maybe spot a value that overall is an improvement on leaving the setup at the default.
 
When I was attempting my calibration it was on a warm still sunny day, with the tripod weighted down by my camera bag on concrete too - I think it was stable ;)
 
Fair enough. Just one thing to eliminate. Not everyone thinks these things through. ;)

I know from my efforts to shoot the moon at 400mm or 560mm while in 10X Live View just how sensitive such lenses are to vibration.
 
My experiences with a 1Ds3 and two 1D4's have been fine with the exception of one lens, the 50/1.2L. It's consistent on both the 1D4's but is nothing other than completely random on the 1Ds3.

Bob
 
I have done a 70-200, a 300 non is and a 200/2 all with and without a 1.4x and 2x.

One or two did vary a bit but eventually settled down to a good amount.

All mine done outside in good light. Target stuck on the shed with cellotape and all at roughly the 50x distance.

With the 70-200 I only did it at 200mm
 
Probably worth a mention that when I did witness erratic behaviour with some lenses the results improved with repeated runs. I can't help feeling that the focus mechanism loosens up a little if worked a bit before testing, perhaps more so if the lens (and lubricants) are cold - like 10c rather than 20c for example.
 
HoppyUK said:
It doesn't do anything that you can't do manually, and seems to be less accurate and inconsistent judging by all those posts on POTN.

I guess it's a fun toy but TBH I don't see the point. If you have a good target and know what you're doing (not hard, and you still need that with this computerised thing) it only takes a few minutes and you know it's right.
Click to expand...

I posted the above six months ago. Is it any closer to being 100% reliable yet?

And does anyone actually trust it to be optimum, without also double-checking conventionally? Which is hardly a difficult task, and therefore begs the question of what's the point?
 
Can't even get a test to complete with the D3S!
Printed the target on coated matte paper and stuck onto a solid box for mobility.
D3S on a Feisol carbon fibre tripod (solid as a rock!).
Just no joy ... tried with 85mm f1.4 and 24-70 f2.8.
Shame as it seems an obvious option to do it in a computerised manner rather than fiddle and readust :(
Waiting on support to come up with a suggestion.



capture61.jpg
 
I have a hunch that this clever idea will never work 100% reliably without having custom profiles for every possible camera/lens permuation.

It's well known that third party manufacturers struggle with reverse engineering of lenses and flash guns. Radio triggers too - ask Pocket Wizard ;) There are always quirky combos that don't play together as they should, and this stuff is changing all the time as new products come out. Even Canon has some strange compatability issues on the 5D3 with a couple of its own lenses.

And given that this seems to be a sledgehammer to crack a nut anyway...?
 
Thats what I had with he 70-300 VC, tried a few times.

When I got in touch he said:

My suggestion would be to change the settings to the following:

- Test Point Consistency Level: None
- Test Point Amount: Many

This will then cause errors in AF to be averages out over many points, rather than trying to catch a single representative AF point each time. There are some lenses (generally 3rd party lenses) which can often behave less reliably than manufacturer own lenses, and the settings above generally sort out the issues.
Click to expand...

In fairness I had to go away for work for a while and never set it back up - but the after sales is excellent.
 
Sounds like a 'make it produce a result' rather than a 'get a correct result' option to me :thinking:
 
When you see the message that the results are not as expected it usually means that the sharpness data is all over the place and the software has no way to be confident in setting any particular AFMA value.

I can only speak for my own results, but when a body and lens are working well and set up in a suitable calibration environment - good light and solid support - you should see a sharpness plot which looks something like this result from my 5D2 and 85/1.8....

20111224_111713_000.png


.... with a pretty much perfect bell curve giving a strong indication of where the sweet spot should be for adjustment.

When you have poor lighting, poor support or poor body/lens consistency you may get results something like this plot from my 5D2 and 50/1.4....

20111224_110001_000.png


.... with the sharpness result plots all over the map. I don't see how the software can be blamed for poorly performing hardware (or setup).

In my opinion one very tangible benefit afforded by the software is that you get to easily see the (in)consistency of your hardware performance. You can see which lenses are working well and which should perhaps not be relied on for perfect focus every time. This kind of thing is far from obvious when trying to laboriously shoot manually adjusted test shots and compare results between one AFMA value and the next.

Even some of my better glass sometimes throws in a spurious result, but by plotting a trend line you (and the software) can look past the odd error and by examining the the bigger picture instead of individual plots determine the best result for that body/lens overall.

Finally, the results illustrated above are pretty repeatable over and over again. The 50/1.4 is always pretty lousy. The 85/1.8 is mostly very consistent. I certainly don't blame the software because my 50/1.4 is practically beyond help. I blame the 50/1.4. I hear the 50/1.8 is as bad if not worse.


EDIT : p.s. When you have three bodies and nine lens to match together, as I do, I think the benefit of automating much of the AF calibration is well worth the price. If you only have one body and a lens or two, perhaps slow ones at that, then maybe it is a sledgehammer to crack a nut.
 
Last edited:
tdodd said:
I don't see how the software can be blamed for poorly performing hardware (or setup).
Click to expand...

I take your point but a Nikon D3S (under 500 clicks) with a Nikon 85/1.4 and Nikon 24-70/2.8 on Feisol heavy duty legs in daylight is not my idea of poorly performing hardware or setup :shrug:
 
gramps said:
I take your point but a Nikon D3S (under 500 clicks) with a Nikon 85/1.4 and Nikon 24-70/2.8 on Feisol heavy duty legs in daylight is not my idea of poorly performing hardware or setup :shrug:
Click to expand...

I take your point too. :) but on what surface is your tripod placed - carpet, floorboards, paving slabs, concrete, earth? Anyone moving around and sending vibrations through the floor? If outdoors is there any wind? Is the tripod extended or collapsed? Are you using mirror lockup, and if so for how long? I have mine set to 2 seconds because I believe that is the bare minimum for vibrations to settle, at least with longer lenses and my tripod and head. If you switch to live view and zoom in as much as possible do you see any evidence of vibration on the screen?

Like I say, I can only speak from my own experience as a Canon shooter and, for me, the software does exactly what it says on the tin and it does it well. Any issues are, in my experience, not a fault of the software. Your mileage may vary.
 
Last edited:
tdodd said:
I take your point too. :) but on what surface is your tripod placed - carpet, floorboards, paving slabs, concrete, earth? Anyone moving around and sending vibrations through the floor? If outdoors is there any wind? Is the tripod extended or collapsed? If you switch to live view and zoom in as much as possible do you see any evidence of vibration on the screen?

Like I say, I can only speak from my own experience as a Canon shooter and, for me, the software does exactly what it says on the tin and it does it well. Any issues are, in my experience, not a fault of the software. Your mileage may vary.
Click to expand...

Inside the front uPVC porch with double glazing and triple polycarb roof (hence virtually clear daylight) with not a hint of a draught and on hogs hair tiles covering a solid concrete floor. The tripod is about 2/3 collapsed and is legs only (38mm dia carbon fibre), no centre column and rock solid! No evidence of vibration and I wouldn't expect any with the setup.
 
Then I'm out of ideas. Have you contacted Rich (the developer)? He was very helpful in the early days whenever I ran into problems or came up with suggested improvement whilst I was beta testing. Although I've had no need for further contact with him since he does appear to be very responsive to support queries.
 
Ah, now I just tried the 100mm - it returned a +1 and spot on focus.

I think I owe the software an apology, but it f does make me wonder how different the quality control is between third party and Canon.

I have just had my 50-500 OS come back from Sigma and now its spot on, before it just always seemed to be that little out - now I've just got a Sigma 70-200 and again I see the same thing.
 
JohnN said:
Thats what I had with he 70-300 VC, tried a few times.

When I got in touch he said:
My suggestion would be to change the settings to the following:

- Test Point Consistency Level: None
- Test Point Amount: Many

This will then cause errors in AF to be averages out over many points, rather than trying to catch a single representative AF point each time. There are some lenses (generally 3rd party lenses) which can often behave less reliably than manufacturer own lenses, and the settings above generally sort out the issues.
Click to expand...


In fairness I had to go away for work for a while and never set it back up - but the after sales is excellent.
Click to expand...

I've tried that change in testing levels but it makes no difference, with both of my D3S cameras and a selection of quality Nikon lenses, the tests always fail before completion.
There is something more going on here than a shaky tripod - I notice that all those having good results are using Canon cameras, maybe this just isn't right for Nikon yet even though it says suitable for D3S :shrug:
Time for a refund I think!
 
Apparently Canon are releasing an SDK tomorrow (whatever that is) which should help this software so it irons out any bugs it has.
 
Can anyone help, I've been struggling with this for the last hour now, getting fed up and have now missed kick off for the rugby match

I have the target search with the green tick but when I click Start on the Fully Auto AFMA, I get an error box saying "Could not get analysis information"

I've been all through both pdfs to see what to do when this occurs but nothing

The lighting is bright, the tripod is stable, the floor is solid tile floor, I've not moved
 
If this is with the 1D3 I occasionally see glitches like this, but usually retrying without actually altering a thing seems to work. If not then maybe a good old fashioned reboot will get things going again. You could also try another USB cable, just in case.

I assume you're on the latest FoCal release. If not then check for updates and maybe even try the latest beta.
 
Yes. Spent the first 10 minutes keep clicking start in the hope it would do something. That's when I started unplugging everything and rebooting. I have now given up and gone to rugby, might make the second half :LOL:

Yes, it is the most recent version. Only bought it last week.
 
kelack said:
Can anyone help, I've been struggling with this for the last hour now, getting fed up and have now missed kick off for the rugby match

I have the target search with the green tick but when I click Start on the Fully Auto AFMA, I get an error box saying "Could not get analysis information"

I've been all through both pdfs to see what to do when this occurs but nothing

The lighting is bright, the tripod is stable, the floor is solid tile floor, I've not moved
Click to expand...

Had that with my D3S's, whatever lens, line up with the target ok, get the green tick then zippo ... can't do it!
 
I didn;t have much luck with it on the 1D3, now for the 5D3 its fully manual so I've not tried more than once, for me a waste sadly.
 
I have found with the latest version it's pretty quick with the 5D3 & 1Dx in Semi-Auto mode.
 
Are there any decent tutorials out there as I've never had much luck with it.
 
JohnN said:
Are there any decent tutorials out there as I've never had much luck with it.
Click to expand...

It's a long time since I've read it, and quite a while since I've used the software, but my recollection is that the documentation supplied was actually very good. It may look long winded and tedious to wade through, but it's actually quite informative and enlightening.
 
Apart from it doesn't say what to do when you get the "Could not get any analysis information" :LOL:
 
I didn't know they were on Facebook. Will have a look when I get home from work. Thanks
 
kelack said:
Apart from it doesn't say what to do when you get the "Could not get any analysis information" :LOL:
Click to expand...

Hi Kelly

I'm Rich (the developer of FoCal). Sorry you're having trouble with the software.

The "Failed to get analysis" message is shown when no other reason can be determined for why the shot couldn't be taken or analysed.

From looking through above I see you're using a 1D3. The 1-series cameras have quite a few custom functions which can stop FoCal running correctly, so it's worth checking a few things:

1. Make sure you don't have AF expansion enabled
2. Make sure you've only got a single focus point selected (although FoCal should pick that up)
3. Make sure that the camera *can* be put in Av mode, ISO 100 and Spot Metering (I know on the 1Dmk4 you can limit the AE modes (P/A/S/M), ISO range and metering modes available so I guess you can on the 1Dmk3 too).

If you haven't already, it's worth sending an email to support@fo-cal.co.uk as it will get into our ticketing system and we can keep track of helping you with this.

I hope this helps.

Rich
 
Hi Rich and welcome to TP :wave:

Thanks very much for replying to the thread. So that message is a bit like the engine warning light in a car then :LOL:

I think I may have had it in ISO 200 when I tried. I was definitely on AV and Spot Metering with centre point focus only. I will have another try this Saturday - it's too dark when I get home from work in the evenings. I will also check that I haven't got AF expansion enabled in the custom functions then

Thanks
 
It really depends what you're using the software for, Dale.

I'll assume that it's for focus calibration so I'll say the widest aperture available. The software will analyse images captured with different adjustment values (essentially moving the plane of focus) and suggest/choose the best fit. Using a larger aperture will give a shallower depth of field and the MA value offsets will become more obvious.

Bob
 
You must log in or register to reply here.
Back
Top