Zillions of adapter chart tests

L

Lenilenapi

Carbonite Member
OK I finally figured out how to get these tests I've been doing on the web.
They compare Brevis, SGPro, Brevis Flip with Brevis, Brevis Flip with SGPro Rev2, and Letus Extreme

They are all archived @:

http://www.leonardlevy.net/Adapter_stills_master_Final.htm

You can see charts displaying the infamous Brevis f1.4 to f2.8 no light loss phenomena and compare to an SG or a Letus.

http://www.leonardlevy.net/HVX%20BrFlp+Brev%20&%20SGPro/HVXop_BFlp+Br_N1,4_20FC.pdf

http://www.leonardlevy.net/HVX BrFlp+Brev & SGPro/HVXop_BFlp+Br_N2,8_20FC.pdf

http://www.leonardlevy.net/HVX Letus/HVXop_Let_N1,4_Mid_24FC.pdf

http://www.leonardlevy.net/HVX Letus/HVXop_Let_N2,8_Mid_38FC.pdf


You can compare sharpness on any of them and look at how well a Brevis works with an SGPro ( I only had a Rev2 but it should be the same if it connects as well.)

You can see the difference between center focus and best compromise "mid" focuses, and see that all the adapters are fine if you take the time to get a Mid focus.

You can see how the Sony Ex-1 works with the first generation of Brevis and Letus adapters before their new achromats.

Look at a Letus EX zoomed in to see if it has a a sharper flatter field.

Compare the difference between DOF on the same lens at the same f stop with 2 different adapter screens:

http://www.leonardlevy.net/DOF/DOF test Inf_Foc_B_SGN1,4.pdf

http://www.leonardlevy.net/DOF/DOF test Inf_Foc_BrN1,4.pdf

You can compare light loss from a chart shot without adapters (on a Naked HVX), to those shot with adapters when wide open.

These are not perfect charts - read the explanation at the beginning. I did these by myself as best I could. I lit them with a light bounced in the ceiling to save time and that did not give the charts even exposure and sometimes screwed up color balances. I sometimes mislabeled charts and then corrected the mistakes by erasing things in photoshop or adding info to the original chart.
The footcandle readings are aproximate only as my meter wasn't digital.

The stills were made by using QT conversion in FCP to make jpg stills at medium compression. They were then transformed a bit more which probably added more compression, so try to use them for the comparison information they show not as accurate images of what you would get in a video image.

These images all looked much brighter and cleaner on my monitor by the way.

Let me know if they are valuable or if I just wasted a huge amount of time.


Thanks.

Lenny Levy
 
Last edited:
The best part of all of 'em is showing the difference between "Auto focus" and "adjusted focus" on the ground glass.

I'm not gonna do charts, but your suggestion definitely helped out on that one.

Also, the deal gets better once you get more light because you can stop the HVX down to 4 - 5.6 for exposure and get even better results.

EDIT: I do have to ask, though... if you were at different "zoom numbers" on the HVX for each one... how did you compensate for each one? Did you move the camera back and away from the chart? move it closer to fill the frame or are you just cropping out of the image?

What spawned the decision to use different zooms instead of setting each one at the exact same? With the Flip unit on it should've been possible to have all setups at the exact same FOV... although I know your preference is max... perhaps all done at 35mm FOV would've been ideal?
 
Last edited:
YAY!!! CHARTS!!!!!!!

what I'm gathering from a couple of these -

- http://www.leonardlevy.net/HVX BrFlp+Brev & SGPro/HVXop_No_Adapter_12FC.pdf
- http://www.leonardlevy.net/HVX BrFlp+Brev & SGPro/HVXop_BFlp+Br_N1,4_20FC.pdf
- http://www.leonardlevy.net/HVX BrFlp+Brev & SGPro/HVXop_Br+Sg_N1,4Mid_24FC.pdf

the brevis flip/50mm 1.4 loses about 1 stop (contradictory to dennis's video... it only lost 0.5 stops in that)
the sgpro flip/50mm 1.4 loses about 1.5 stops.

would you agree with that or am i screwing something up?

thanks
patrick
 
p.s. with the "HVX no adapter shot", did you zoom all the way out? cos that could equate for the brevis' missing 0.5 stops

thanks again
patrick
 
jenningsp said:
the brevis flip/50mm 1.4 loses about 1 stop (contradictory to dennis's video... it only lost 0.5 stops in that)
the sgpro flip/50mm 1.4 loses about 1.5 stops.
Click to expand...

I wouldn't get too hung up on the exact numbers - they are aproximate. A spectra pro has exact numbers at 4, 8, 16, 32, 64 Footcandles etc. Everything in between is a guess and even the angle of the meter matters. Plus I was alone bounmcing a light off the ceiling and trying to get a waveform to the same place everytime.

I'd put the loss as somewhere between 1/2 stop and 1 stop on my tests. Thats doesn't mean Dennis's were wrong. They're close and probably used different methods.
 
Kholi said:
I'm not gonna do charts, but your suggestion definitely helped out on that one.

Also, the deal gets better once you get more light because you can stop the HVX down to 4 - 5.6 for exposure and get even better results.
Click to expand...
If you don't bother to do your own charts, how do you know whether your adapter improves as you stop down to 5.6? Maybe it reaches its best at 2.8. Maybe your better off stopping down your 35mm lens. Your just guessing and 30 min with chart would replace that with knowledge.

I wanted to test stopping down the video cameras, but with 3 adapters and 8 lenses , enough was enough for now. I will check it out later. It would be a nice test for someone to do though. I was happy to see how well the Nikon @ 1.4 performed on the EX @ 8 .

Kholi said:
I do have to ask, though... if you were at different "zoom numbers" on the HVX for each one... how did you compensate for each one? Did you move the camera back and away from the chart? move it closer to fill the frame or are you just cropping out of the image?

What spawned the decision to use different zooms instead of setting each one at the exact same? With the Flip unit on it should've been possible to have all setups at the exact same FOV... although I know your preference is max... perhaps all done at 35mm FOV would've been ideal?
Click to expand...

They are different units optically and mechanically with the screens at different distances , often using different achromats - there is no reason to expect them to function properly at the same zoom and focus settings on the HVX. I basically zoomed in to the first place I had a useable image without vignetting and for the Brevis & SG that seemed tvery close to what my Nikon SLR saw at the same distance. I moved the camera in and out a teeny bit to adjust it though.

I did a few tests here and there at what i estimated to be Academy and that involved moving the camera back to still fill the chart. I don't personally give a wig about Acadamy though.
Vingetting on the Letus made me move back further than the SG & Brevis generally. The important thing was to fill the chart. I don't think I was always perfectly centered on the chart BTW which could account for different corners being slightly different.

Also because of the bounce light the lower corners are generally darker. At a certain point I got brighter footcandles by moving my bounce closer and that made the fall off even more of an issue. Threre was only so much I could do though and I got too much specular going direct in the small room I was in.

Lenny
 
Lenilenapi said:
If you don't bother to do your own charts, how do you know whether your adapter improves as you stop down to 5.6? Maybe it reaches its best at 2.8. Maybe your better off stopping down your 35mm lens. Your just guessing and 30 min with chart would replace that with knowledge.

I wanted to test stopping down the video cameras, but with 3 adapters and 8 lenses , enough was enough for now. I will check it out later. It would be a nice test for someone to do though. I was happy to see how well the Nikon @ 1.4 performed on the EX @ 8 .
Click to expand...

Because, knock on wood, my eyeballs haven't degraded to the point where I can't tell when softness is becoming an issue. I can also clearly see when the camera's connected to a 37 and 42" panel when things are getting "sharper?. Now, make sure you note the keyword: issue. It has never been on my Letus until I put a crappy 17-35mm on my unit. I still haven't seen the insane edge softness you got here. We're not about to go round about it, though.

You already know it gets better when stopping down. It's been that way with adapters since the M2 was first released.

They are different units optically and mechanically with the screens at different distances , often using different achromats - there is no reason to expect them to function properly at the same zoom and focus settings on the HVX. I basically zoomed in to the first place I had a useable image without vignetting and for the Brevis & SG that seemed tvery close to what my Nikon SLR saw at the same distance. I moved the camera in and out a teeny bit to adjust it though.

I did a few tests here and there at what i estimated to be Academy and that involved moving the camera back to still fill the chart. I don't personally give a wig about Acadamy though.
Vingetting on the Letus made me move back further than the SG & Brevis generally. The important thing was to fill the chart. I don't think I was always perfectly centered on the chart BTW which could account for different corners being slightly different.

Also because of the bounce light the lower corners are generally darker. At a certain point I got brighter footcandles by moving my bounce closer and that made the fall off even more of an issue. Threre was only so much I could do though and I got too much specular going direct in the small room I was in.

Lenny
Click to expand...
To me it seems like if your position changes, distance from the chart, then the "apparent" edge softness changes depending on where your zoom is set.

If all cameras could've gone to Academy and sat there EVEN IF YOU DON'T CARE ABOUT ACADEMY then the test would've been a bit more equal. Instead, there's now another variable to consider: Camera distance from the board.

This is the only reason I would even consider printing out a chart to test my Letus... because of preferences. You don't want to use 35mm FOV, there are others that won't dare go outside of 35mm FOV.

Since all adapters may not be able to do it, and you being a one man band, it sounds more logical to me to have set the zoom equal at all points at least on the HVX. I.E leaving it at 35mm FOV.

Either way, looks like lots of fun.
 
Last edited:
I think that the HVX's natural softness covers up the softness of lenses. Edge to edge is a different story, but keep in mind that a 35mm still photo lens may be soft at f/1.4 if you're printing film but for the HVX it's hardly noticeable.

What I wanted to know is if you had a spacer ring on the Brevis sans flip unit because the edges look soft. I've never had edge to edge issues like what I see on your charts.


Marco
 
marco0782 said:
I think that the HVX's natural softness covers up the softness of lenses. Edge to edge is a different story, but keep in mind that a 35mm still photo lens may be soft at f/1.4 if you're printing film but for the HVX it's hardly noticeable.

What I wanted to know is if you had a spacer ring on the Brevis sans flip unit because the edges look soft. I've never had edge to edge issues like what I see on your charts.


Marco
Click to expand...

Another good point to keep in mind. HVX is naturally soft for one.

I think that using a still lens on the adapter will also expose the still lenses flaws more so because it's just not made for moving video. It's a "hack" for those who can't afford or don't want to rent the real deal. SuperSpeeds and Cookes are world of difference altogether. The image clarity is about as immediately apparent as strapping on an 80-200/2.8 Nikon. Only slightly better. =D

As far as the Brevis and it being soft on the edge: Maybe Lenny is right in that you won't see the edge softness until you do a chart? I said the same thing about the LetusEx in that what I saw Tim and Lenny experiencing was NEVER an issue with my unit. I only saw it with my 17-35mm Tamron @ 17mm.

Sooo maybe it IS time for us to go and shoot a chart?
 
Marco, No I didn't use a spacer ring on the Brevis and have only seen you recommend it as i recall. i don't think Dennis recommends it so i never tried it. You may well be right that improves things though. Where do you put it?

Its pretty immaterial to me at this point because I see no reason not to use the Flip, but it wouyld be interesting to test. I was a bit surprised at the corners on the original Brevis, and actually reshot those tests to double check. I didn't notice it at all (esp the CA) on my 17" Panasonic monitor.

Kholi - I already explained why I think having the same zoom setting does not make the images optically equal in any way at all.

Re: 35mm acadamy. I didn't have the time to place each adapter on a wall with an 8'5 x 11 sheet of paper 2' away to determine their individual academy zoom settings. If Academy is important to you than I highly recommend you do those tests. I'd love to see them.

My guess is you would find that the optics of some adapters might be better at full SLR than Academy and vice versa. If you do those test I recommend shooting both. There are a few test of Academy here just to see if they were way different.

Re: how good anyone's eyeballs are at figuring these things out without a chart. From the contradictory information I've read on these pages - I'd say pretty piss poor.

I note that you had suggested multiple times that zooming in on the Letus Ex would give much better cornering and overall focus. I tried that and got strikingly opposite results. Maybe your tests would be different though.

http://www.leonardlevy.net/HVX Letus/HVXop_Let_N1,4_Ct_Z39(Zoomed-In).pdf

http://www.leonardlevy.net/HVX Letus/HVXop_Let_N1,4_Ct_24FC, Z25.pdf

I don't think these tests are perfect, but for issues of sharpness they tell you a whole lot more than anecdotal shots of autumn in the woods. Not as much fun to shoot though and certainly not as pretty.

Lenny
 
Lenilenapi said:
Marco, No I didn't use a spacer ring on the Brevis and have only seen you recommend it as i recall. i don't think Dennis recommends it so i never tried it. You may well be right that improves things though. Where do you put it?

Its pretty immaterial to me at this point because I see no reason not to use the Flip, but it wouyld be interesting to test. I was a bit surprised at the corners on the original Brevis, and actually reshot those tests to double check. I didn't notice it at all (esp the CA) on my 17" Panasonic monitor.

Kholi - I already explained why I think having the same zoom setting does not make the images optically equal in any way at all.

Re: 35mm acadamy. I didn't have the time to place each adapter on a wall with an 8'5 x 11 sheet of paper 2' away to determine their individual academy zoom settings. If Academy is important to you than I highly recommend you do those tests. I'd love to see them.

My guess is you would find that the optics of some adapters might be better at full SLR than Academy and vice versa. If you do those test I recommend shooting both. There are a few test of Academy here just to see if they were way different.

Re: how good anyone's eyeballs are at figuring these things out without a chart. From the contradictory information I've read on these pages - I'd say pretty piss poor.

I note that you had suggested multiple times that zooming in on the Letus Ex would give much better cornering and overall focus. I tried that and got strikingly opposite results. Maybe your tests would be different though.

http://www.leonardlevy.net/HVX%20Letus/HVXop_Let_N1,4_Ct_Z39(Zoomed-In).pdf

http://www.leonardlevy.net/HVX Letus/HVXop_Let_N1,4_Ct_24FC, Z25.pdf

I don't think these tests are perfect, but for issues of sharpness they tell you a whole lot more than anecdotal shots of autumn in the woods. Not as much fun to shoot though and certainly not as pretty.

Lenny
Click to expand...


You always get different results compared to everyone elses. Ain't just me, but now it's Marco, it's also the people who said the SGpro loses more light than Letus. Your results were different.

I also said Brevis is soft on the sides and I could see that clearly with my eyeballs, along with other people. You said you couldn't.

So, in the end, I'm gonna assume that anything I test is going to pretty much turn out differently than what you're doing for whatever odd reason. =D

But, I and I'm sure a lot of other people thank you for showing us the importance of not just using Auto Focus for focusing in general.

You still confuse me on the zoom setting. If you have to move the camera forward til it's 1 foot from the wall to get the chart to fit... doesn't that change how far the edge softness extends into the chart itself?

A 50mm 1.4 lens with HVX @ 24 will have a wider FOV than a 50mm 1.4 with HVX @ 39 ... so either you have to crop the image in post to compensate or you move the camera. It changes variables, does it not? That's all I'm saying: Your variables aren't constant.
 
Thanks for the vote of confidence Kholi. If you want different kinds of tests do them yourself. If you can't see any value in these I'm sorry.

Re: Framing - I moved the camera. Cropping would invalidate everything.
And I admit as I did tests I sometimes shot with same adapter very slight distances differently. i don't think they were significant. I think I kept the important variables relatively constant but you don't - OK.

But a Letus or a Brevis or an SGPro at a specific zoom setting on the HVX are going through different achromats onto units with their GG different distances away from the HVX. So the same zoom distance for all three is simply irrelevant.

re: the chart on the wall 2' away - That's how Barry tells you to figure out what Academy is. How do you do it?

Re: Luis's tests: The HVX is at f5.6 and above. I expect all of these guys to do well there, but I usually am shooting in much lower light. That's why I tested with everything open. that's where you will see the limitations. I would have loved the time to shoot at higher f stops as well.

Actually Luis' test mostly shows you that the 2.8 Nikon is sharper than the 1.8 and nearly up to the limit of HVX resolution. He should crop his shots though so that the black border is gone. Then you get corners farther into the corner and the resolution numbers are accurate.

I like Luis's tests as a general rule. I've learned from them.


Lenny
 
Last edited:
8.5 x 11" paper, 2 feet from the image plane in the adapter, crop 1/4 of the sides off.

On an HVX the measure is Z27.3mm for 35mm FOV. Tim Naylor found the exact same number, Barry_Green did also.
 
And, Hey, remember the "use one camera" variable? Turned out that using two different cameras got different results, so using just one was the best way to keep the variable constant.

I'm just sayin', same thing to me. Might not be in the end, but it would seem that they should all be tested at the same exact FOV so there's no moving the camera and changing that variable. S'ALL i'm sayin.
 
Lenilenapi said:
Re: Luis's tests: The HVX is at f5.6 and above. I expect all of these guys to do well there, but I usually am shooting in much lower light. That's why I tested with everything open. that's where you will see the limitations. I would have loved the time to shoot at higher f stops as well.

Actually Luis' test mostly shows you that the 2.8 Nikon is sharper than the 1.8 and nearly up to the limit of HVX resolution. He should crop his shots though so that the black border is gone. Then you get corners farther into the corner and the resolution numbers are accurate.


Lenny
Click to expand...

Check the thread, I believe it was to find out about edge ot edge sharpness in general. As his first test really didn't look anywhere near as nice.

http://www.dvxuser.com/V6/showpost.php?p=1124195&postcount=1

It pretty much looks like yours. However, if that's THE case in the test-- that you stated-- then that works. Ugh. I don't want to do any res charts but seems like it'd be ideal to just do it at the same settings.

P.S. =P YOu told me I was just guessin' about the image getting better at lower HVX stops. But, you now say you expect it to be like that.

I'm just standin' on your shoulders, here. Teasin', ya know?
 
Barry ,
I must have a mental block about this academy framing thing.

Please clarify

You put up a piece of 8.5 x11 - Horizontally or Vertically? (The paper itself is clearly not 16:9 right.)

You frame it 2' away - but do you make the horizontal or vertical side of the video image match the paper?

Which side do you crop off 1/4" from - horizontal or vertical? (Couldn't you cut it off to begin with?)



"On the HVX its Z27.3"

With what adapter? Are you so sure they are all the same. They sure don't look that way to me though I could be wrong.

I'm sure that the HVX gets sharper with the adpaters as you stop down. But is the effective value of that complete by f2.8, f4, f 5.6? That's what i would want to know. I've seen some people in all seriousness say that you shouldn't use an adapter below f8 on the video camera.


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