LensRentals.com

A Bit of 3250mm Fun

Posted by

It's winter. It's slow. We get bored. We have lots of stuff to play with. Too much stuff and too much time leads to all kinds of trouble. I mean experimenting to find combinations of equipment that might benefit our customers. So anyway, Joey decided it would be a superb idea to take a tiny little Nikon V1 mirrorless camera with it's 2.7 crop factor and put it on the back of a Nikon 600mm f4 VR with a 2X teleconverter. I mean, who doesn't need a 3250mm equivalent lens, right?

Joey's Frankenphoto system

It was kind of overcast and gray and there's not a lot of scenery behind our building (this is the view shot at 35mm). Joey's telephoto test target is the cell phone tower on the right, the smaller one.

The scenic view from Lensrentals loading dock.

To avoid getting you too disoriented, this is the top of the cell phone tower, shot with a 600mm f4 VR on a Nikon D3x.

The cell phone tower we use as a telephoto test target, shot at 600mm

The white pole in the center is where we focus (it's kind of lost in the gray sky, but if you look down from the top antenna you'll see a black bird sitting on top of it. And here's the view with the V1 at 3250mm.

The V1 with 600mm, 2x converter (3250mm effective view)

Of course, the image above is compressed down to 800 pixels wide to fit in this blog post. Below is a crop at 100% of the area along the green wire, lower right 1/3 of the image.

Yeah, it's a little grainy and noisy, that small sensor is no D700. But scroll back up to the first image, the one shot at 35mm. It's a rather impressive bit of magnification, isn't it?

For another example (because we believe anything worth doing is worth doing to excess), here's the view looking the other way from the loading dock, again a 35mm image on a full frame camera. Notice the two doors above the metal stairs in the center of the image.

Zoom in with a 600mm lens on a regular camera and you can tell the door number. Oooh-aaah.

But slap the V1-2X-600mm combination on there and you can read the alarm company's phone number. All of you would-be Paparazzi should be thinking "cellulite-celebrity-on-the-beach-pictures from half a mile away".

Of course, we could get nearly as much magnification by cropping the images from a D3X shot at 1200mm. But this was way more fun.

23 Responses to “A Bit of 3250mm Fun”

Brad said:

What sort of wiring job is that on the cell tower? Must be how they make that 4G so fast.

Roger S. said:

At the end of the month, I hope to buy one of your used lenses. Not because it's a good deal (which I think it is), not because we're both named Roger, not because I'm originally from Tennessee, and you're in Memphis (which is culturally north Mississippi, where I live now).

I intend to buy it from lensrentals.com because of this great blog, which I have spent hours reading this week, and because of the bats&*t crazy fun y'all have doing stuff like this and the chess set. Thanks for being awesome!

James Shepherd said:

Sloppy wiring using tiewraps--That lens/camera combination would be a delight for a project inspector--I also noticed the tape coming off a connector... (15 years of working on top of towers finally mixes with photography)

LensRentals Employee

Roger Cicala said:

Roger, you used "Memphis" and "culture" in the same sentence. I believe there are rules against that sort of thing :-)

Best,

Roger

Roger S. said:

I've lived in small town Mississippi for almost 19 years. Culture is relative.

Samuel Hurtado said:

it would be interesting to check, but I'm sure you'd get A LOT more detail on the v1 than on a crop from a big sensor camera

that V1 is like a crop from a 75mpix full frame camera

and looking at dxomark, now I know why I don't want such a super-high-mpix camera: noise may or may not be averaged out when resampling that theoretical 75mpix to 16mpix (another interesting thing to check), but in any case DR is greatly reduced; assuming V1's sensor uses tech as least as good as the D5100, dxomark tests say this small photosite design leads to nearly 3 stops less DR
http://www.dxomark.com/index.php/Cameras/Compare-Camera-Sensors/Compare-cameras-side-by-side/%28appareil1%29/745|0/%28brand%29/Nikon/%28appareil2%29/698|0/%28brand2%29/Nikon/%28appareil3%29/680|0/%28brand3%29/Nikon

Jay Frew said:

Roger ~ Thanks for another entertaining read.

The other Roger said "...and you’re in Memphis (which is culturally north Mississippi, where I live now). "

You could also mention that the three biggest cities in Mississippi are (in no particular order):

- Memphis;

- New Orleans; and

- Mobile

;~))

Cheers! Jay

Carl said:

Great job Roger and Joey!

Samuel, great info!

What does the AF do with the V1 on these lenses, or does it work at all? Sorry if that is too ignorant of me!

When spring/summer comes, go to a beach somewhere and try the test again...you can guess at the subject matter I would rather look at, haha!

I saw a similar test (as have most of you I'm sure) with a 400 f/2.8 and the 2x TC via a link from nikonrumors, of shoes through a window in a shoe store. I have to say I'm amazed that the lens sharpness holds up so well with such tiny pixels, mostly because of the attached TC. Without the TC, I bet these super-teles could resolve pixel detail vastly smaller than the V1's...albeit with less lens magnification or effective focal length (than by using a TC).

I notice nobody seems to be trying the V1 on the Sigma 800 f/5.6...I presume it must be too soft even without a TC...as compared to a Nikon super-tele, with one attached? Or is it just that many rental places don't stock any of the Sigma lenses longer than 300mm? Maybe someone will try a V1 on the Canon 800mm with 2x TC. Or better yet, on a telescope with the same retail price as the 800 !!! Of course, I guess a scope couldn't focus closely enough on subject matter to truly test the resolution, without atmospheric conditions getting in the way.

I am really looking forward to the D800. If it can come close to, or beat, the noise of the D3x (compromised though that one is), but with 36 mp rather than 24...That would really be something (and cost less than half as much!!). It would basically be the same pixel "site" size as my Canon 50D crop camera. Best of all would be if the D800 could shoot RAW at lower resolution without necessarily going into crop mode, as Canon cameras can do. Some reviewers tend to slam the "RAW1, RAW2" feature, but it's ideal if you know you want to shoot at the highest ISO and don't want to waste memory on pixels that are just all noise anyway. Since the D800 can go to ISO 25k, I can see where a feature that "combines" pixels would be welcome. And sometimes, my sDOF shots feel like I am wasting full resolution when so much of the frame is intentionally out of focus anyway...

This is where you say "no, the noise will be worse"...!!

I was comparing Nikon zoom lenses at thedigitalpicture, and in the particular samples tested, it appears the 28-300 actually is sharper everywhere in the frame at all focal lengths, to the corners, (wide open) than the 24-120 f/4. So the 28-300 would probably be the lens I would buy...if I got a Nikon camera. The distortions seem easily corrected. The CA actually also looked worse on the 24-120, and even on the 24-70 f/2.8...than on the 28-300! Of course Nikon can correct the CA in-camera...I assume even on the raw NEF files (?). Even if it didn't, that is one of the more fun and quick things I like to correct in PS.

I still think Canon is a better company overall, though...but that's just me! I just wish they would have made a "1D Mark 5" with 25 or 28 MP on a full frame (or else kept the "h-crop" as a variant or something)...rather than trying to beat Nikon at their own game with lower noise and bigger pixels (with the 1Dx). But, I can see why most of the market wants the 1Dx, and to go full-frame. And if it has to be the sports camera also, then obviously there can't be a high pixel count...nor would it be necessary (if it somehow magically could have both).

I guess I am the oddball (as usual), because it seems many D700 fans definitely don't want a lot of pixels. They (mostly) liked things how they were! But times will always change. I feel bad for them; they don't want to have to pay a lot more and buy the D4!

Samuel Hurtado said:

This V1 seems like the perfect tool for trying to bring in some facts to the "more megapixels vs fewer megapixels" debate

I'd put it against the D3s (it should be similar sensor technology, if anything the comparison would be biased towards the high megapixels camp, as the sensor in the V1 is more modern) (D3x or D700 could be used instead of D3s, but that gives even more advantage to the high-megapixels camp, as they are even older bodies)

I'd choose a fairly standard, fairly sharp lens, probably a 50mm

and I'd run two tests:

* detail and diffraction: take pictures from same spot and lens, with all bodies, varying aperture and shutter speed, and compare 100% crops from each to see if the virtual 75mpix fullframe sensor actually delivers more detail than the fullframe 12mpix sensor and how (if?) that narrows down when diffraction kicks in

* ISO and noise: take pictures from same spot and lens, with all bodies, at f/4 or so, varying shutter speed and ISO; compare crops from the D3s with downsamples from the V1 (so the final pictures show the same area of the subject and have the same pixel count), to see if downsampling actually averages out the extra per-pixel noise from the virtual 75mpix sensor

(in both cases, maybe throwing in the virtual 37mpix fullframe sensor from the D7000, and the actual 24mpix sensor from the D3x, could be a good idea)

looks like the perfect task for a bored LensRentals employee :)

LensRentals Employee

Roger Cicala said:

Carl,

We talked abou the Sigma 800 but really think at these distances vibration reduction may be beneficial, even on a tripod. Maybe not though. But the Nikon lenses do maintain autofocus with the adapter on the V1 and J1, although it's slower than on a Nikon SLR body.

Stephen said:

Can I work there please? Dang, I would love to play with gear and tell people what I think about it. I'm jealous Roger, very jealous.

Ben said:

Actually, I use this technique all the time and with great results. I shoot video for a living and I've got a Panasonic HPX-300 HD video camera. The sensor on this camera is 1/3 of an inch diagonally. When I attach my Nikon 300mm f 2.8 with the TC20 III it produces a focal length equivalent of approximately 4000 mm. That may sound like an awful lot, but it allows for some really spectacular wildlife shots (assuming the air is still and heat waves are at a minimum.} Not to mention frame-filling sunsets!

Elliott McCrory said:

You need to do this test on the Sigma 200-500mm f/2.8. If the Amazon.com reviews are any guide, you'll be able to see all KINDS of interesting things with the V1/2X/SigmaMonster combination.

Bryan said:

that is a beast! that said i regally connect my camera to my Celestron CPC800 8" diameter telescope with a focal length of a little over 2m plus the 1.6 crop on my canon makes for a pretty long lens but not very portable.

Samuel L said:

Bit of a false comparison, as a Celestron C8 (which is what's being used there) doesn't throw a 44mm image circle which is what would be required to fully illuminate a 135-format sensor, making the "crop factor" multiplication at best a half-truth. As an amateur astronomer, doing stuff like this seems so obvious and everyday...

Haruo Chikamori said:

~jaw dropping on floor and shattering~. OMG!

Hey, if you want to go birding on a budget...this is the best way to do it. Slap on a 70-300mm with the V1 lens adapter and you get a 810mm lens. And here I thought that I should get myself a 600mm f/4 lens for my D300s. ~evil grin~

Antonio said:

a-maze-ing lol

touristguy87 said:

This is all very well and good but really not too useful as in my experience it's usually better to just move closer and shoot without the TC. Of course it does come in handy when you don't really have another choice. Maybe it'll save some Saudi prince from ordering another 2600mm Canon lens.

You should try it at night and see if you can get any interesting shots of the near planets :)

touristguy87 said:

ps, forget about the whole "virtual 75mm fullframe" thing and just treat it as it is: a 10MP 1" diagonal cmos sensor...and downsampling by definition doesn't average-out noise, it just discards the extra pixels. So the SNR improves by the downsampling ratio (20log10 of it). And comparing the V1/J1 to the D5100 doesn't mean that the smaller pixels cost up to 3x the DR. It means that the D5100 is a better camera. Nothing more. Try it with the Sony A200, a 10MP CCD subframe. That's why they do these tests instead of just generalizing from pitch.

touristguy87 said:

...a comparison against the NEX-7 and the A850 is also informative, though unfortunately they don't have the relevant data for the D3200...anyway this is one reason that people pay extra for Nikon gear

Tord S Eriksson said:

I own the V1, and a K-5, and NEX-5N, while the wife has a OM-D E-M5 instead of the V1, otherwise we have the same set-up.

As long as the light is good, the V1 is superb, with the fastest focus in town, but as light gets dimmer it is not nearly as crisp as the OM-D5, so cutting down the OM-D's digital images to V1 size gives a lot of interesting facts: The OM-D is so pre-sharpened that PP doesn't do much good, unless the exposure is off (a rare event), while the other cameras take to PP in the same way!

A few examples, made with the K-5, OM-D and the V1:

http://www.flickr.com/photos/tord55/7177393223/in/set-72157630043299435/

Ken said:

I've managed to create a similar setup with a V1 + Questar - also about 3200mm equivalent focal length. Manual focus and no VR but works well on a heavy tripod. Extension tubes and eyepiece projection could make it even longer but that would be just crazy lol

Leave a Reply