Will the NN4 support a 5D MkII and 24-70 lens?

[Frankster]

Concerning RD16 there is only one fault with 200mm focal length: there is no 4º set up. It would be 3.75º with 45%.

An overlap of around 30% is very good. Normally we say that 25% is enough. Overlaps >=45% may cause stitching errors, cause the program might generate too many control points. But to get to know, you should make some test shots.

As the upper rotator - M1_L - is concerned, I can not follow your thoughts. Actually you have to calculate the degree of Pitch = how many degrees you move the upper rail up or down and how many rows you have to shoot to get 180º in vertical.

Hi Heinz,

My bad with the 4° figure, thanks for pointing that out. Unfortunately degrees got rounded along with the percentage figures and I forgot to reformat them in Excel to show decimals. The underlying value was ok at 3.75, it just displayed the rounded value 4. BTW Same thing happened with the 8°, should have displayed as 7.50. They're displayed correctly below.

If overlaps >=45% may cause stitching errors, then it appears I'm going to have problems due to the 7.5° limitation with the upper rotator.

Perhaps you could not follow my thoughts on the M1-L upper rotator because of the wrong percent overlap figures that John pointed out. Or maybe it's because I used the number of click stops instead of degrees like I did in the RD-16 figures. I see where you've translated a +45° pitch to 6 clicks of the upper rotator. My thinking is not to use degrees for the upper rotator in my table because it all boils down to using click stops anyway. Yes/No? Better to keep it simple in the field and just put the click stops in the table instead. No math, no mistakes, less work.....

I've updated the overlap figures for the M1 using the formula John offered.

(FOV less pitch degrees) / FOV = % overlap

Is that the formula you would use? If so, and if the new figures look realistic to you, then the overlaps are so confining that I don't think the M1 is going to work for me. First of all, there is no acceptable overlap at all for the 135mm lens. And if I limit the results to only those where the overlap percent falls between 25% and 45% (instead of 25% and 49%) then the 70mm lens is of no use to me to either.

@zoom 24 you would need 3 rows at +45º, 0º, -45º plus Z/N shots. Upper rotator always clicks by 7.5º per steps, so to reach a pitch of +45º, you have to click 6x.

Plus you have to calculate that using longer focal lengths and getting to 8 rows for example, you can reduce the number of shots coming closer to Zenith or Nadir.

But the upper rotator's 7.5º will be ok . I only have to recalculate my NN5 settings with its 2.5º steps for new pitch steps.

I think I understand what you're saying about the number of rows and 180°. Problem is - I'm not interested in 180°. Making mosaics of scenes originally framed with a 24mm or 28mm lens is what I'm after, which would require focal lengths of 35mm and higher. The 24mm and 28mm lenses are in the table simply because I have them, not because I intend to use them for multi-row 180° panos. I do not understand how the upper rotator's 7.5° will be ok given the overlap figures shown in the table.

Perhaps I should reread this thread and look harder at the NN5.

Hope you get your clamp tomorrow.

Frank, I checked your calculations for the percentage overlap for 70mm. The horizontal figures are ok, but the vertical ones appear to be wrong. For a pitch increment of 15°, the percentage overlap is (28.84-15)/28.84x100=48%. For a pitch increment of 22.5°, the percentage overlap is (28.84-22.5)/22.84x100=22%.

John

John,

Thank for pointing that out. If one was wrong, then the whole column was wrong. I used a formula I wasn't sure about, yours is much simpler and does seem to make more sense. The updated table using your formulas is shown below. There's quite a difference though, and some limitations I didn't expect. Do these figures look more realistic? For your perusal I also included the complete table for the M1 figures, which includes all the overlap figures I didn't use that were under 25% or over 49%.

Regards,
Frank

Table for camera mounted in PORTRAIT orientation (overlap figures rounded to the nearest integer.)

	RD-16					M1-L Number of Click Stops (@7.5° each)
	-------------------			---------------------------------------
24mm		30°	36°			5	6	7
		43%	32%			49%	39%	29%
								
28mm		24°	30°			5	6	
		48%	35%			43%	31%	
								
35mm		20°	24°			4	5	
		47%	36%			45%	31%	
								
50mm		15°	18°	20°		3		
		44%	33%	26%		43%		
								
70mm		10°	12°			2		
		48%	38%			48%	<--Smallest usable overlap ?	
								
135mm		6°	7.50°					
		41%	26%				<--No usable overlap ?	
								
200mm		3.75°	5°			1		
		45%	27%			27%		


	7.5°	15.0°	22.5°	30.0°	37.5°	45.0°	52.5°
	1	2	3	4	5	6	7   <--click stops
24mm	90%	80%	69%	59%	49%	39%	29%
28mm	89%	77%	66%	54%	43%	31%	20%
35mm	86%	72%	59%	45%	31%	17%	4%
50mm	81%	62%	43%	24%	5%	-14%	-33%
70mm	74%	48%	22%	-4%	-30%	-56%	-82%
135mm	51%	1%	-48%	-97%	-147%	-196%	-246%
200mm	27%	-46%	-119%	-192%	-264%	-337%	-410%


Basis for the above figures.....
Field Of View for D700 36 x 23.9 sensor
Lens mm
24mm	73.74°	52.94°
28mm	65.47°	46.22°
35mm	54.43°	37.70°
50mm	39.60°	26.88°
70mm	28.84°	19.38°
135mm	15.19°	10.12°
200mm	10.29°	6.84°

[John Houghton]

... there is no acceptable overlap at all for the 135mm lens.

It's perfectly possible to use the 135mm lens if you should want to. The guidance that Heinz offered ("Overlaps >=45% may cause stitching errors, cause the program might generate too many control points) does not clearly indicate what the actual potential problems are. You cannot merely have too many control points. In PTGui Tools->Options you can set your own limits on how many control points are generated. The trouble with excessive overlap is that you can get control points generated between non-neighbouring images. In theory, this ought not to be a problem, and often may well not be, but experience shows that restricting control points to the horizontally and vertically adjacent images generally gives the easiest trouble-free stitching. This just means that you need to exercise some control over the control points generation to ensure that you only have control points in the positions that you want.

Another possible problem with excessive overlap concerns blending. With more than 50% overlap, some blenders may completely discard the central image from three overlapping images as being surplus to requirements, since complete coverage will still be obtained with no hole left. All this means is that the possibly poorer quality image edges will be used instead of the better quality from the centre of the discarded image.

But whatever the problems, there's a simple and effective way to avoid them: just crop or mask the unwanted image areas away to give <50% overlap!

John

[nick fan]

I'm located in the eastern US, any reason why I couldn't order from you instead of a reseller?

It seems the only complication is getting the unit with the 210mm upper rail instead of the standard 170mm rail. Could this be done easily with an online order, or would it be best to phone in the order?

you just contact the people at nodalninja.com
Live chat or phone call or email.
I don't do any retail sales. I only develop and make NNs. No time for retail sales.

Nick

[Frankster]

John,

Thanks for your feedback. I do understand how one wouldn't want 3 adjacent images sharing the same control points, but thought that would only occur if overlap was 50% or more. This is why I've limited my overlap options to between 25% and 49%. However, I now realize we're teetering on the edge of tolerances/consistency the closer we get to the 50% mark, and that it's likely one could use what he believes to be a 48% overlap for example, but ends up with 3 images containing the same control points in spite of that.

I suspect Heinz's experience in this regard tells him to build in some "wiggle room" to avoid such borderline situations, and thus tries to limit the overlap at <=44% just for peace of mind of knowing there won't be any unanticipated surprises in the stitching program. That's worth knowing as I never thought it would be an issue, though it does narrow my options when it comes to choosing an overlap for a given focal length.

Being locked into 7.5° increments with the M1, and looking at my table, makes it clear just how limited my choices are with the M1. I like the modular concept and the M1 seems the most versatile unit, except for the upper rotator's inflexible 7.5° increment. I do use PTGui Pro and admit to being a novice at using it. Having said all that, it seems the best approach here is the same for most photography - "get it right in-camera" to avoid post-processing headaches down the road. Would you agree?

Regards,
Frank

[Frankster]

Thanks Nick Fan. Perhaps you or Heinz could respond to this...

Originally I favored the 7.5° M1 click intervals over the 15° NN5 intervals. But now that I can pre-visualize how it works by looking at the table, it seems the 15° NN5 click intervals weren't so bad after all. Anywhere the table shows an even number of M1 click stops, I could theoretically halve that number and apply it while using the NN5. The big difference being that the NN5 gives me the option to fine tune the overlap but at the expense of losing the more convenient click stops.

Now that I've re-focused on the NN5 because of the way its upper rotator works, I have renewed interest in some of the points that were touched on earlier in this thread.

Heinz said,

M1 has 7.5° built in upper rotator steps. If you like to be free to set up even xx.2,5° steps on the upper rotator, and be free to use lenses up to 70-200, a very good alternative would be the NN5 irr RD16 for nearly the same price as NN4.

OK, sounds better this time:-) By "steps" do you mean manually adjusting the pitch by using the markings engraved on the rotator, or do you mean using actual 2.5° click stop intervals instead of the 15° stops? I've looked at the NN5 manual and didn't see anything about changing the upper rotator click stop intervals. Could you please explain or supply a link?

From one of the images Heinz posted, it appears I can get the NN5 with an Arca style clamp on the upper rail. And given what was already discussed about that clamp and the RRS L-plate, I would want to order with that clamp and try it out. However, I can't find where we ever discussed what it would take to use a RRS clamp on the NN5 as it seems we got sidetracked with the M1 overlap stuff. My apologies for any confusion this caused.

What would it take to use the RRS B2 LR II clamp on the NN5 upper rail?

Heinz, since you are about to receive this clamp and prepare it for the M1, perhaps you could also comment on fitting it to the NN5.

I think I remember seeing or reading that it can't be attached directly to the upper rail due to the NN5 rail's "U" shape. And though the NN Arca clamp appears to fit, I suspect its rear is machined to fit the same way the Camera Mounting Plate (part#NCP-1) fits, see #9 on pages 7 & 8 of the NN5 manual. If so, what parts would be needed to mount it to the rail and would the NN5 be able to tolerate the added thickness and still be used to mount the camera L-plate in both portrait and landscape orientations?

I believe the RSS clamp itself would add 11mm of thickness (after allowing for the dovetail to fit into the L-plate).

Regards,
Frank

[John Houghton]

Having said all that, it seems the best approach here is the same for most photography - "get it right in-camera" to avoid post-processing headaches down the road. Would you agree?

Yes, I think that is sound advice, but one should also be aware of the capabilities and limitations of the software products you are using, and learn how to use them properly. PTGui has very few limitations and it's not at all difficult to stitch images with 90% overlap if you need to. In fact, I stitch images with 100% overlap from time-to-time, and also with 0% overlap too.

John

[Frankster]

I wasn't aware PTGui gave me so much leeway. Thanks for sharing that, it will be good to know when the time comes.

Regards,
Frank

[hindenhaag]

Frank,

I will receive the clamp this afternoon. Several things went wrong with the post. Anyway, I hope it will be delivered to the post office this afternoon.

Now let me tell you about some principles of NN5 upper rotator. Actually you can fix it to 2.5° pitch intervals manually using the marks engraved on the degree ring. There is no click. Additionally, you have a lever, with which you can set an extra 15° pitch set up which will move in place by using the lever. This will secure heavy weights in place and it moves in place without looking to the marks: 0°, 15°, 30° and so on. Without using the lever you can set the pitch manually by using 2.5° ring marks. Using the lever, you move the upper rotator with open rotator to for example 30°, lock the lever which moves it into place and fix the knob.

The M1 upper rotator, once opened, may be clicked by 7.5° intervals, the upper rotator of NN4 by 15° degree intervals. You cannot use intervals in between the clicks.

On NN5 you have to add the NN Arca Style Clamp on the upper rail and directly use your L-Plate with it.

Cause I felt you are more on M1, I left the NN5 discussion. I got your questions and will try to answer them later today when I got the RRS Clamp in my hands.

Regards,
Heinz

[Frankster]

Heinz,

I've watched the NN5 video on using the upper rotator with the lever, so I have a good idea how it works. I like having the option to move the lever so I can manually set the pitch. It seems very handy for those times when 15° click stops gives me either too much or too little overlap.

I don't like the M1-L because its upper rotator is fixed at 7.5° intervals and I cannot manually set the pitch.

I don't like the NN4 because I can't make full use of all the focal lengths with my D700 and Nikor 24-70 f2.8 lens. And gathering from what you just said, it's fixed at 15° intervals and cannot be set manually - same reason I don't like the M1-L.

We'll talk more about the clamping once you have a chance to work with your new RRS clamp. I've read this thread in the NN5 forum.
The equipment discussed is exactly what I've had in mind, especially the clamping. His final post (just before yours) lists all the parts, among them is the RRS clamp he attached to the NN5 upper rail. He used a B2-40 LR whereas your new clamp is a B2 LR II. In addition to yours being 10mm longer, it also has more holes drilled through it and most appear to be threaded. Do you think your B2 LR II clamp would be better suited to the NN5, or do you think the poster chose the B2-40 LR for some functional purpose that cannot be achieved with your B2 LR II clamp?

Sorry about the confusion between the M1 and NN5 in the discussion, I didn't realize 7.5° intervals would be so inflexible until I saw the actual overlap percentages for my lenses. I knew that interval would give more overlap control than 15°, but didn't expect it would still leave me wanting for more. There's so much to consider, it's hard taking it all in at one time.

Regards,
Frank

[hindenhaag]

Frank,

I got you with all the needs for equipment you like to reach for personal use.

I got the RRS B2LRII lever clamp cause I needed another clamp for my own needs using different tripods, NN Hardware, and NN pole based on my personal decision. Besides this, I offered the chance to you to test it for your personal questions to use this special clamp you have been asking for use on M1.

Now you talk about NN5. I understand that it might be complicated to get to know about the different upper rotators on NN Models. This is why I try to help people. Right now suddenly you are asking for a different RRS Clamp model B2-40 LR.

The forum tries to help people to get them on the road. But it won't work, asking for different equipment within days. You have to inform yourself on the internet for example on the RRS website to think about a lever clamp that would fit your personal needs. And then ask about this model.

So I will try my RRS clamp on both M1 and NN5 for you and let you know about it. And then I am out of this discussion, and this is the first time since years I state this. As I recommended a time ago, you can rent both NN models, and I think you have to buy both of the RRS Clamps to try them on your own.

Regards and success,
Heinz

[nick fan]

Sorry about the confusion between the M1 and NN5 in the discussion, I didn't realize 7.5° intervals would be so inflexible until I saw the actual overlap percentages for my lenses. I knew that interval would give more overlap control than 15°, but didn't expect it would still leave me wanting for more. There's so much to consider, it's hard taking it all in at one time.

there are lots of choice between 70 and 135mm. if you use zoom, you can use any focal length in between. Just put a small label on the lens to remember it in the future. If you use primes, there are 85, 90, 100, 105mm. Our NN Ultimate M2 will support finer intervals. so you will have more control on the overlap. But I am also sure it will introduce new problems and limitations.

NIck

[Frankster]

Heinz,

I already did ask RRS via email which clamp to use before making my first post here, and they said to use the same B2LRII that you just got. BUT they don't know what it takes to MAKE their clamp work on NN products, such as drilling/tapping a hole for the antitwist screw on M1 for example. I always try to inform myself as much as I can before asking people on forums to spend their time answering my questions.

I think you'd agree that informing myself would include reading the NN forums. The forum thread I pointed you to seemed to contradict RRS's advice and I thought you might know if there was a good reason why the B2-40 LR was chosen over the B2LRII for the NN5. That's all I asked and I am sorry it caused you so much frustration. I already use the B2-40 LR on my ball head, it's a very common clamp. I know you have a great deal of experience, and I know you have a RRS L-plate. I just thought you might also have this common RRS clamp and would know the reason off the top your head without doing any further work. Apparently you don't have it, so again I apologize for causing you so much frustration.

But I am frustrated too. First to find that the NN4 won't work with my camera/lens, and then to find the M1 upper rotator's fixed pitch could not give acceptable overlap with my lenses. Never having used a panohead I expected it would be complicated finding a model that would work. I know the RRS panohead will work in every respect (overlap & clamping) but chose to search for a brand that has click stops because they appeal to me. I never expected click stops on the upper rotator would be such a problem. Perhaps I'd be better off getting the RRS pano head to use on a NN RD-16 rotator, but I'll wait to see what you have to say before taking that route.

I do appreciate the time you've spent and look forward to your comments about mounting the RRS B2LRII clamp on the NN5.



Nick,

Thanks for your feedback.

...or I could use the NN5 and have the options to either manually set the upper rotator OR mark the lens with tape. I prefer having that versatility.

The reason is, that above 70mm the only other option I have is 135mm. Although it does come from a 35-135 Nikon push/pull zoom, it's an older lens of the non-CPU type that doesn't record focal length in the EXIF data. There are no markings on the lens between 70 and 135mm, so I intended to use it fully extended at 135mm should I want to use it for a mosaic. That's the only way I can be certain of the focal length. Moving it anywhere between 70 and 135 without having EXIF data to confirm the focal length setting for labelling is a crap shoot, and would result in way too much guesswork both in finding the best overlap and in the stitching program that would want me to input the focal length manually because it cannot read it from EXIF data.

Frankly, I'm getting pretty worn out trying to find the right NN model from those available. To rehash the same concerns for yet another model, one that doesn't seem to be available, is a bit too much for me at the moment.

Regards,
Frank

[nick fan]

I think you'd agree that informing myself would include reading the NN forums. The forum thread I pointed you to seemed to contradict RRS's advice and I thought you might know if there was a good reason why the B2-40 LR was chosen over the B2LRII for the NN5.
I do appreciate the time you've spent and look forward to your comments about mounting the RRS B2LRII clamp on the NN5.

what do you have? both clamps will work.

The reason is, that above 70mm the only other option I have is 135mm. Although it does come from a 35-135 Nikon push/pull zoom, it's an older lens of the non-CPU type that doesn't record focal length in the EXIF data. There are no markings on the lens between 70 and 135mm, so I intended to use it fully extended at 135mm should I want to use it for a mosaic. That's the only way I can be certain of the focal length. Moving it anywhere between 70 and 135 without having EXIF data to confirm the focal length setting for labelling is a crap shoot, and would result in way too much guesswork both in finding the best overlap and in the stitching program that would want me to input the focal length manually because it cannot read it from EXIF data.

exif data is not needed in stitching. Just put a mark on the lens which gives you desirable overlap. Then zoom the lens to this mark each time you want to make the pano.

Nick

[Frankster]

what do you have? both clamps will work.

Nick,

I only have one clamp but that is already used on my ball head and will stay there, it's the B2-40 LR. I would need to buy another clamp for the NN.

As you probably know, the B2LRII is 10mm longer but unlike the B2-40 LR the center hole is threaded and there are other threaded holes as well.

In the NN5 manual at the top of page 15 it shows the camera mounting plate NCP1 with a screw meant to screw into the camera bottom, but in my case I suspect it would screw into the RRS clamp. This implies that the RRS clamp should be threaded like the camera is, and that the B2LRII would be suitable clamp.

In the thread I pointed Heinz to, you made this post#8 where you recommend a B2 LLR II 80mm LR clamp w/ dual mount. This clamp is virtually identical to the B2LRII in that it has threaded holes, it's just longer than the B2LRII. So far everything makes sense.

However, because the clamp you first recommended was sold out, in this post#13 you recommended the B2 AS II clamp, which does NOT have a threaded hole.

And in this final post#23 the original poster listed the clamp he ended up attaching to the camera plate on the NN5 upper arm. It was a B2-40 LR, which also does NOT have a threaded hole.

As well, the NN RP1-14 low profile rail plate was discussed in that thread, but you said it wasn't needed. And indeed it was not in OP's final list of parts.

So if you wonder why I'm confused it's because
a) I can't see how an unthreaded clamp can have a screw screwed into it. And
b) though I do see in the manual one threaded hole in the NCP1 camera plate, it's obviously used to hold the plate to the upper rail.

The only way I could see that both clamps could work, is if the hole in the camera plate meant to fasten the plate to the camera has a 3/8" threaded hole through which the 1/4" camera screw passes without using the threads.

If that is not the case, how could both clamps possibly work if one is the threaded and the other is not?

exif data is not needed in stitching. Just put a mark on the lens which gives you desirable overlap. Then zoom the lens to this mark each time you want to make the pano.

Nick

Maybe PTGui doesn't need the EXIF data but it does want to know focal length, and for that I'd have to guess.

Regards,
Frank

[nick fan]

In the NN5 manual at the top of page 15 it shows the camera mounting plate NCP1 with a screw meant to screw into the camera bottom, but in my case I suspect it would screw into the RRS clamp. This implies that the RRS clamp should be threaded like the camera is, and that the B2LRII would be suitable clamp.

And in this final post#23 the original poster listed the clamp he ended up attaching to the camera plate on the NN5 upper arm. It was a B2-40 LR, which also does NOT have a threaded hole.

As well, the NN RP1-14 low profile rail plate was discussed in that thread, but you said it wasn't needed. And indeed it was not in OP's final list of parts.

You can remove the 1/4" screw on the NCP1/2. Then you have a 1/4" socket for mounting a QR clamp with counter sink.
But if you want to use the low profile rail plate, you need a clamp with thread at the center.

Maybe PTGui doesn't need the EXIF data but it does want to know focal length, and for that I'd have to guess.

focal length has no meaning for Ptgui. It only cares angle of view of lens which is calculated by optimization.


Nick