Date   

Re: SONY RX100 synch measurements

George Themelis
 

Let’s think about this for a second.

 

What kind of synchronization does 1/1000s represent?

 

Let me ask you this question: When you go out shooting, how often to you have to use a shutter speed higher than 1/1000s to freeze the action?

 

Answering for myself: Rarely. Normally, I use my cameras in P mode. If there is action that I need to freeze, I usually use Shutter priority mode and 1/500s or 1/1000s. This freezes just about everything. The pictures of the dog were taken at 1/500s shutter speed.

 

If 1/1000s shutter speed can freeze the motion in my picture, then 1/1000s synchronization will work.

 

Yes, I am fully aware that shutter speed and synchronization are two different things, but they are related through this very simple logic: If 1/1000s speed freezes all movement, then 1/1000s synchronization does not create any perceivable mis-sychronization.

 

So, for what I shoot, 1/1000s shutter speed freezes the motion and 1/1000 s synchronization does not create any perceivable mis-sychronization.

 

The RX100 camera tested by Werner showed 1/1250s average synchronization and this is called only sufficient? In what world is this happening? Certainly not in mine. If you are shooting specifically high speed sports photography then maybe this is not good for you, but for the rest of us shooting casually in nature, city, where there is some movement, certainly frozen at 1/1000s, then this is great news.

 

George

 

From: George Themelis via groups.io
Sent: Monday, April 19, 2021 12:13 PM
To: main@photo-3d.groups.io
Subject: Re: [Photo-3d] SONY RX100 synch measurements

 

“Well these results are by far not perfect, not even excellent - I would say the synch result is only 'sufficient'”

 

Well, I am sorry but anything at 1/1000s or higher is excellent for me.  Here we are looking at almost 1/2000s. This is better than the Samsung NX1000 and many other twin rigs that people have been happily using.

 

These are great results, thank you for confirming!

_._,_._,_

 


Re: SONY RX100 synch measurements

George Themelis
 

“Well these results are by far not perfect, not even excellent - I would say the synch result is only 'sufficient'”

 

Well, I am sorry but anything at 1/1000s or higher is excellent for me.  Here we are looking at almost 1/2000s. This is better than the Samsung NX1000 and many other twin rigs that people have been happily using.

 

These are great results, thank you for confirming!

 

PS. For the past few days I have been trail-running with a dog and taking action shots. I photograph the dog running at full speed and jumping over logs. The dog is “frozen” mid-air over the log.

 

If the cameras can capture this kind of action, they can capture anything else that I shoot, so I am very pleased. Very-very pleased. I just came back from another dog walk/run photoshoot, and, as I was looking at today’s pictures, I was thinking how pleased I am with these cameras. Did I mention that I am pleased?  I don’t care how the rest of the world feels but I am the only one that I need to please, and this is happening. 😊

 

George

 

 

From: Werner Bloos
Sent: Monday, April 19, 2021 11:37 AM
To: main@photo-3d.groups.io
Subject: [Photo-3d] SONY RX100 synch measurements

 

Hello,

SONY RX100 cameras paired for 3D seem to be popular and their synch is claimed to be 'perfect' or 'excellent' - but no one did synch test by now to prove this synch. Let me destroy this illusion: I have done synch tests with a pair of RX100M2 and got these results:
- individual power-on for each camera: median missynch is: 1/1254s (= 0.797650 ms)
- synchronised power-on via the MULTI Terminal port: median missynch is: 1/1698s (= 0.589034 ms)
(of course a power-off/power-on was done for each shot to eliminate synch drift)

The synchronised power-on improves synch but with this model not as significant as with other models (e.g. DSC-HX50)

Well these results are by far not perfect, not even excellent - I would say the synch result is only 'sufficient'
OK, now someone else may say: 'The M2 model is old and my newer model is much better...'  - ok then provide your synch tests with newer models (M3-M7) to be able to compare them...

You can find my spreadsheets as well as all results here:
https://docs.google.com/spreadsheets/d/1PVmr8RjcRJvL1kEas5lHL4TQZqeQBixhwKBuO73HyIM/edit?usp=sharing

greetings
Werner



 


SONY RX100 synch measurements

Werner Bloos
 

Hello,

SONY RX100 cameras paired for 3D seem to be popular and their synch is claimed to be 'perfect' or 'excellent' - but no one did synch test by now to prove this synch. Let me destroy this illusion: I have done synch tests with a pair of RX100M2 and got these results:
- individual power-on for each camera: median missynch is: 1/1254s (= 0.797650 ms)
- synchronised power-on via the MULTI Terminal port: median missynch is: 1/1698s (= 0.589034 ms)
(of course a power-off/power-on was done for each shot to eliminate synch drift)

The synchronised power-on improves synch but with this model not as significant as with other models (e.g. DSC-HX50)

Well these results are by far not perfect, not even excellent - I would say the synch result is only 'sufficient'
OK, now someone else may say: 'The M2 model is old and my newer model is much better...'  - ok then provide your synch tests with newer models (M3-M7) to be able to compare them...

You can find my spreadsheets as well as all results here:
https://docs.google.com/spreadsheets/d/1PVmr8RjcRJvL1kEas5lHL4TQZqeQBixhwKBuO73HyIM/edit?usp=sharing

greetings
Werner



Drones for 3D

davidgaha
 

Hi, I hope everyone is doing fine, and healthy, I would like to know if someone is using a drone, or a couple of them, to shoot stereo images, the possibilities seem to be really good. 
Keep safe 


Re: Phantograms In Virtual Reality

 

On Sun, Apr 18, 2021 at 11:41 PM, JackDesBwa|3D wrote:
If you are using a browser that supports WebXR (e.g. Firefox Reality, Occulus browser...) and select this mode in the viewer, the phantograms on Stereopix are displayed horizontally in the virtual reality environment.
 
Yes, I know, Jack, and I also use the WebXR mode, but this is a whole 'nuther monkey in my zoo.
I'm using side-by-side view (crossed or parallel) in Stereopix (or any other image visible on my computer screen(s)), and the OVR Toolkit converts them to VR mode on-the-fly. The advantage to this is I can look at these *inside* virtual reality environments via SteamVR.
This is, by far, the kewlest thing I've ever seen in 3D, so I'm working on 3D pop-ups (phantograms) in VR, next, and I have a choice of using traditional phantograms, or actually creating them inside VR. I need to create some of my own, since I'm going to show this process on YouTube (in 3D), and don't want to use copyrighted images.


Re: Too little deviation?

gl
 

On 19/04/2021 08:54, Matej Bohac wrote:
Tom, your statistical evaluation is very interesting, thank you for this!

But as you write, the deviation reported by SPM's autoalignment cannot be fully trusted. I don't really understand why, but my experience is that this number rarely corresponds to what I come up with by measuring manually, and many times it is absolutely off.
Having coded my own auto-alignment code, one reason is that the code attempts to match similar features in each image, and calculates deviation from those matches only.  but not everything matching in both views in real life produces a match in the algorithm.  They have limitations generally (things like image noise), and they will miss hard-to-match features like dense foliage or low-contrast things like smeared clouds or horizons.

So it's quite possible for the code to miss features that are closer or further away than the matches it has already found, and therefore underestimate the minimum/maximum deviation in the image.  And there's also the risk that it has mis-matched a feature and created a bogus deviation value from it.

So yeah, don't trust the deviation readouts blindly, they may often be ballpark, but other times way out.
--
gl


Re: Too little deviation?

robert mcafee
 

Of possible interest to the topic of adjusting for different viewing platforms 

“3D Perception Adjustment of Stereoscopic Images Based Upon Depth Ma”. Jong In Gil, Seung Eun Jang, & Manbae Kim Dept of Computer & Communication Engineering Kangwon National University Chunchon Republic of Korea 

From Y S Ho (editor) PSIVT 2011 part 1 LNCS 7087 pp 13-21 Spinger Verlaine 2012




On Sunday, April 18, 2021, 12:44 PM, robert mcafee <geargod2@...> wrote:

I was just reading a technical publication about a method using depthmaps that would automatically adjust the deviation depending on the display method (phone, desktop display, cinema screen).

Bob


Re: Too little deviation?

Oktay
 

I had written an article two years ago about the technical aspects of stereo photography.

This article which includes my hand drawn drawings is so far published nowhere.

Attached is this article where I present first time my views about subjects like deviation, lenses, separation, viewing methods, etc. , which are the subjects of this thread.

It may be a little long and may require previous knowledge of some basic optical and geometrical concepts but I am sure reading it would be interesting among some of you who want to explore deep into the subject and who may want to know about other people's opinions and methods.

I will welcome and appreciate your comments.

Oktay


Re: #3d-cameras #3d-cameras

Bill Costa as just a member
 

The electronic viewfinder is nice but you will not get the normal spacing or the use of flash of the NX1000. That’s something to consider.


Funny -- I like having the option of doing flash photography, yet I can't tell you the last time I actually wanted to use one, even indoors -- with any of my cameras.  On the other hand, it's a rare outdoor outing where I haven't wished I had a viewfinder because I was shooting blind due to glare on the camera display.  The time I do like using the display for framing is when holding the camera over my head or low to the ground -- I imagine it would be nice to have an articulated display for this.  (As of yet I do not have such a camera.). Of course the larger interaxial on this Sony twin would be a minus for close to the ground shooting.
 
...BC

--
Bill.Costa@...
+1.603.435.8526
https://mypages.unh.edu/wfc
No good deed goes unpunished.


Re: Phorlorn Phereo Phailed

JackDesBwa|3D
 

The renewal of the SSL is a formality (someone just needs to submit some documentation) or there is a cost involved?

The certificate was signed by Let's Encrypts, which provides the service for free thanks to automated verifications and large tech company support.
Depending on how it was configured on the server of phereo, it could be as simple as running a command, or something more time consuming (stop the server, run the challenge, copy the new files, restart the server...). Anyhow, it has to be done by the maintainer.

On Stereopix, I configured this to be fully automated and managed by the server itself [it took some efforts to make it work fine autonomously], so that I do not forget to renew it every 2 or 3 months, even if Let's Encrypts sends mails a few days before the expiration date.

Perhaps the last person from Phereo who was doing some maintenance might renew.

I hope so.

JackDesBwa


Re: Too little deviation?

robert mcafee
 

I guess I mistook deviation for how the window was set (if course a crop and realignment sometimes changes the deviation)


On Mon, Apr 19, 2021 at 7:49 AM, George Themelis
<gathemelis@...> wrote:

Reprocessed how?

 

Both of my images were one of a kind action shots. I cannot go back and reshoot them. It’s one thing if this is a tabletop or something I can reproduce and shoot again and another if that was a once in a life-time opportunity.

 

George

 

 

From: robert mcafee via groups.io
Sent: Monday, April 19, 2021 7:42 AM
To: main@Photo-3d.groups.io
Subject: Re: [Photo-3d] Too little deviation?

 

George might even post the image as iriginslly submitted for competition and then reprocessed with more disparity (and maybe less) and have people vote which version they liked best


Re: Too little deviation?

George Themelis
 

Reprocessed how?

 

Both of my images were one of a kind action shots. I cannot go back and reshoot them. It’s one thing if this is a tabletop or something I can reproduce and shoot again and another if that was a once in a life-time opportunity.

 

George

 

 

From: robert mcafee via groups.io
Sent: Monday, April 19, 2021 7:42 AM
To: main@Photo-3d.groups.io
Subject: Re: [Photo-3d] Too little deviation?

 

George might even post the image as iriginslly submitted for competition and then reprocessed with more disparity (and maybe less) and have people vote which version they liked best

_._,_._,_

 


Re: Too little deviation?

George Themelis
 

Thank you guys for the input.

 

Regarding the question of why the SPM deviation cannot be trusted is I related to how SPM gets this number.  SPM uses a number of points to base the alignment on. I believe that the maximum deviation is based on these points only. If the nearest or farthest objects are not in this group of points, then the deviation will be different than measured.

 

I think it is always better to measure the deviation  because then you have the choice to select which points you think are important and this involves some degree of judgement.

 

George

 

From: Matej Bohac
Sent: Monday, April 19, 2021 3:55 AM
To: main@photo-3d.groups.io
Subject: Re: [Photo-3d] Too little deviation?

 

Tom, your statistical evaluation is very interesting, thank you for this!

 

But as you write, the deviation reported by SPM's autoalignment cannot

be fully trusted. I don't really understand why, but my experience is

that this number rarely corresponds to what I come up with by measuring

manually, and many times it is absolutely off.

 

Back to the original question George posted:

 

I often bracket the base when taking close up shots with a beamsplitter

macro-box. So I end up with a series of images with various deviations

to choose from.

 

My personal experience is that close-ups of flowers and similar objects

often look more natural with a much smaller deviation than 1/30 in

projection. Many images I project have as little deviation as 1/60 or

1/80, even if I had a whole range of versions to choose from.

 

So if I were a judge, I would definitely not consider 1/72 or 1/42 to be

a problem by itself. The resulting impression depends on the object and

composition.

 

Matej

 

 

 

 

 


Re: Too little deviation?

robert mcafee
 

Perhaps George might share his image about which the judge commented only 1/72 disparity.  George might ask for Photo-3D members to vote whether they liked the image (with respect to deviation/disparity only) or whether they thought the image would benefit from greater disparity.

Geroge might even post the image as iriginslly submitted for competition and then reprocessed with more disparity (and maybe less) and have people vote which version they liked best


On Mon, Apr 19, 2021 at 3:55 AM, Matej Bohac
<matej@...> wrote:
Tom, your statistical evaluation is very interesting, thank you for this!

But as you write, the deviation reported by SPM's autoalignment cannot
be fully trusted. I don't really understand why, but my experience is
that this number rarely corresponds to what I come up with by measuring
manually, and many times it is absolutely off.

Back to the original question George posted:

I often bracket the base when taking close up shots with a beamsplitter
macro-box. So I end up with a series of images with various deviations
to choose from.

My personal experience is that close-ups of flowers and similar objects
often look more natural with a much smaller deviation than 1/30 in
projection. Many images I project have as little deviation as 1/60 or
1/80, even if I had a whole range of versions to choose from.

So if I were a judge, I would definitely not consider 1/72 or 1/42 to be
a problem by itself. The resulting impression depends on the object and
composition.

Matej






Re: #3d-cameras #3d-cameras

George Themelis
 

The electronic viewfinder is nice but you will not get the normal spacing or the use of flash of the NX1000. That’s something to consider.

 

Also note that Generation 1 does not use the multiport so it cannot be synchronized. So you are looking at generations 2-7.

 

George

 

 

From: Bill Costa
Sent: Monday, April 19, 2021 1:58 AM
To: main@photo-3d.groups.io
Subject: Re: [Photo-3d] #3d-cameras

 

My recommendation is the Sony RX100 .. * You have the option of several generations (from 2 to 7)

This Digital Photography Review page looks like a good resource for picking out which generation would be a good fit. I have been very happy with my Samsung NX1000 rig with the ring mount, but I have to admit I am getting really tired of not having an electronic viewfinder. The fact that all but the first generation RX100 has one is making me consider going down this road.

...BC

 


Re: Phorlorn Phereo Phailed

robert mcafee
 

The renewal of the SSL is a formality (someone just needs to submit some documentation) or there is a cost involved?

As I recall there was some effort to pass the site over to someone who might maintain it. Perhaps the last person from Phereo who was doing some maintenance might renew. If there are costs involved perhaps this community might help with this.


On Mon, Apr 19, 2021 at 2:50 AM, Stereopix Net
<contact@...> wrote:
I believe the issue I am having with viewing has something to do with flask player. Perhaps my antivirus/firewall blocking?

As I said before, it is because their SSL certificate expired ("Valid from January 11, 2021 to April 11, 2021") and was not renewed.
Then the browser assumes that it is risky and refuses to load the images. 
 
Is anyone else able to access Phereo and view images on their site?

If you force your browser to accept the certificate, it would work. Perhaps if you change the date of the computer too (not tested).
Phereo official Android app, PhereoRoll3D, https://phereo-unofficial.stereopix.net/ are places where you can see the images despite the certificate error.

JackDesBwa


Re: Too little deviation?

Matej Bohac
 

Tom, your statistical evaluation is very interesting, thank you for this!

But as you write, the deviation reported by SPM's autoalignment cannot be fully trusted. I don't really understand why, but my experience is that this number rarely corresponds to what I come up with by measuring manually, and many times it is absolutely off.

Back to the original question George posted:

I often bracket the base when taking close up shots with a beamsplitter macro-box. So I end up with a series of images with various deviations to choose from.

My personal experience is that close-ups of flowers and similar objects often look more natural with a much smaller deviation than 1/30 in projection. Many images I project have as little deviation as 1/60 or 1/80, even if I had a whole range of versions to choose from.

So if I were a judge, I would definitely not consider 1/72 or 1/42 to be a problem by itself. The resulting impression depends on the object and composition.

Matej


Re: Phorlorn Phereo Phailed

JackDesBwa|3D
 

I believe the issue I am having with viewing has something to do with flask player. Perhaps my antivirus/firewall blocking?

As I said before, it is because their SSL certificate expired ("Valid from January 11, 2021 to April 11, 2021") and was not renewed.
Then the browser assumes that it is risky and refuses to load the images. 
 
Is anyone else able to access Phereo and view images on their site?

If you force your browser to accept the certificate, it would work. Perhaps if you change the date of the computer too (not tested).
Phereo official Android app, PhereoRoll3D, https://phereo-unofficial.stereopix.net/ are places where you can see the images despite the certificate error.

JackDesBwa


Re: Phantograms In Virtual Reality

JackDesBwa|3D
 

Okay, kids, if you wanna see something kewler than kewl, look the phantograms at http://Stereopix.net in VR with the OVR Toolkit.

If you are using a browser that supports WebXR (e.g. Firefox Reality, Occulus browser...) and select this mode in the viewer, the phantograms on Stereopix are displayed horizontally in the virtual reality environment.

JackDesBwa


Re: Ingenuity on Mars in 3D #stereopix

JackDesBwa|3D
 

Is it undergoing update, JackDesBwa, or hopefully not some problem with my browser/computer?

The problem was introduced with the update of the viewer on Stereopix (redundant command removed, but still used by this app).
It was fixed a few days ago, but the service which hosts it keeps restarting the app with an older version :-(
I will destroy and recreate the configuration in hope it would fix this.

JackDesBwa

1121 - 1140 of 130507