The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Dear Sony: I want Sensorshift Mulitshot - FW Upgradeable or as an app ?

Stefan Steib

Active member
Dear Folks here is a german article by Traumflieger (with image samples comparing an EM5 II, Canon 5Dsr and a Sony A7r

the sensorshifted MFT chip really shines on 40 Mpix, can anybody imagine what would happen if the A7RII would do the same stuff - 80 to 100 Mpix in brilliant quality ?
I would even pay for an App, I would even pay substantially for such an app - e.g. 100 € or even 200 € ?

Sony do you listen: this should be easy. Please do it !

High Resolution: Vergleich Olympus EM5 II vs. EOS 5Ds / Sony A7R - Traumflieger.de
 

Knorp

Well-known member
Really wonder if the key to sensor shift is to be found in firmware alone.
I recall it could not be done for the EM1 ...
 
V

Vivek

Guest
Really wonder if the key to sensor shift is to be found in firmware alone.
I recall it could not be done for the EM1 ...
Even where they can (the A7 series) do an update (uncompressed RAW) nothing happened. :(

Unless dprevs take up the issue they will not listen (as claimed by a Sony prod manager). They are what they are.
 

CSP

New member
Even where they can (the A7 series) do an update (uncompressed RAW) nothing happened. :(

Unless dprevs take up the issue they will not listen (as claimed by a Sony prod manager). They are what they are.
i can guarantee you this is not true they are very active in collecting feedback, something i never experienced with canon for example.
 

Brian Mosley

New member
i can guarantee you this is not true they are very active in collecting feedback, something i never experienced with canon for example.
The problem seems to be, that they apply that feedback in developing the next model of camera - rather than fixing (or upgrading) things for existing users.

Unfortunately, this fits in with the general perception I have of Sony in terms of customer support. I guess this follows naturally from a 'consumer' brand.

It's good for one's sanity to bear this in mind when buying Sony products - make sure that you want exactly what you're buying.

Kind regards

Brian
 

CSP

New member
The problem seems to be, that they apply that feedback in developing the next model of camera - rather than fixing (or upgrading) things for existing users.

Unfortunately, this fits in with the general perception I have of Sony in terms of customer support. I guess this follows naturally from a 'consumer' brand.

It's good for one's sanity to bear this in mind when buying Sony products - make sure that you want exactly what you're buying.

Kind regards

Brian
...so when did canon improve an old model with firmware the last time ? i have bought 8 1d and 5d series cameras in the last 10 years and i´m not aware of 1 countable improvement beside some minor fixes. and when it comes to feedback they give a **** about your opinion when you don´t devoutly cheer them.....
 

ohnri

New member
The problem seems to be, that they apply that feedback in developing the next model of camera - rather than fixing (or upgrading) things for existing users.

Unfortunately, this fits in with the general perception I have of Sony in terms of customer support. I guess this follows naturally from a 'consumer' brand.

It's good for one's sanity to bear this in mind when buying Sony products - make sure that you want exactly what you're buying.

Kind regards

Brian
The latest round of Sony firmware updates for the A7r2 and A72 just blow this perception out of the water.

I don't ever recall getting anything close to what Sony has done with firmware updates from any of my Nikon gear. Nikon just likes to come out with a whole new body, e.g. 810, 610, with small but important upgrades to fix their old bodies.

So, not only do Sony's latest upgrades push forward existing models, when Sony does release a new body it is often a big upgrade.

I'll take Sony's practice over Nikon's any day.

Current Sony shooter. Former Nikon D4, D800 owner.

But, to stay on topic - I expect there are patent issues preventing Sony from implementing sensor shift multi shot.

But, I would LOVE it if they can add it to my A7r2. Seriously, who cares about hard drive space or file sizes? I need the option of more of everything.

-Bill
 
V

Vivek

Guest
i can guarantee you this is not true they are very active in collecting feedback, something i never experienced with canon for example.
I hope you are correct. In that event, I have a list ...

In addition to the FW updates for the original A7 series, FW updates for RX1 and R and a new RX1s with a built EVF and Tilt LCD and high ISO capacity in a lower MP count sensor( hope this will improve battery life). thank you very much! :)
 

dmward

Member
It seems that the sensor movement required to implement IBIS means that the firmware has the ability to control sensor movement with sufficient precision to make the Sensorshift Multishot function work.

This capability would be extremely useful on both my A7RII and A7II. Even if there were some restrictions.
 

Annna T

Active member
The interest of Olympus (with its small MFT sensors) to find a workaround in order to get more resolution is evident.

However, I wonder about the need to implement such a technique on bodies already having high resolution. If one prints for billboard size, then people have to step back in order to encompass the full picture. So I strive to understand that need for more than 40-50 MB resolution.

And anyway, for the few times that is needed, one can always stitch. It has become easier than ever. I can imagine a huge panorama in a long hallway, where people going by could look at all the details.. But we can already achieve that very easily nowadays. I tried it with a 90mm TSE lens the other day : three rows of seven pictures for a 180 degrees view : one with rise, one without movement and one with fall. Given the level gauge available on the LCD, I could do it with a table tripod setup on the windowsill; I just used a great deal of overlap to compensate the lack of a pano head. It was merged to panorama in LR6 without any flaw. But the resulting dng image weight 1.2 GB. It could be printed to 3.5 meters long. It is amazing how many little details and slices of life one can see in such a picture, but what is the point ? I just did it because I could. To make that thing even more absurd, here is the web version :

Sion and the Rhône Valley by rrr_hhh, sur Flickr
 
Last edited:

ggibson

Well-known member
I tend to think that it is an interesting feature on paper, but in practice the current iteration in Olympus cameras has limited or narrow applications. For outdoor use, almost any wind creates movement between frames resulting in very difficult-to-correct image errors in those areas. When I look back at my photos, there are very few where I think I would be able to apply it. However there is certainly a niche of photographers shooting inanimate objects in a studio who would love this technology.

I certainly think we'll see more of this type of technology to improve image quality in the future though. There will be increases to sensor readout speeds and camera CPU power which will make it possible to stack multiple frames for resolution/noise/color improvements. This type of thing is possible on a desktop now, so it is only a matter of time until it fits into your camera. They just have to figure out ways to overcome the limitations like hand-shake and movement within the shot. Probably a combination of faster hardware and smarter algorithms will eventually be able to compensate these problems. I imagine smartphones are probably where we'll see this type of thing first. Their small sensors can do full sensor readout very quickly and have ample processing power compared to the average camera.
 

Stefan Steib

Active member
Well - why do you think Sinar and Hasselblad are selling multishot backs if nobody needs them ?

These are mainly used for extremely detailed productshots, reprography of e.g. cultural heritage, paintings and plans.

Architecture Photography for the travelling photographer - remember stitching when not doing a flat stitch will require nodal point operation, usable foreground and many more nuisances.
A Sony A7RII can shoot 5 images a second, so for a 4shot that´s less than a second. a 16 Shot could happen in a bit more than 3 seconds !
Thats also definitely faster than stitching avoiding ghosting of moving subjects (a bit) better than doing it manually (which also holds a source of human failures forgetting a shot etc.pp.)

Pushing such a body into this range of gear would blow the market into a whole new ocean of opportunity.
Sure Blad and Sinar won´t like this, but this is what is going to happe anyway. So I´d preferre sooner, than later.

In short - this would be a killer feature - probably Sony will do it in an Alpha 9R Pro Body, maybe together with USB3 internal and external data bus and a bigger battery ? :)

Listen to me Sony - here you got the lawn and it needs mowing badly.
Give us a mower !
 

dmward

Member
The comments about art and architecture are the two areas where I would find this capability immediately beneficial. A combination of higher resolution for a total image and also the ability to crop into details when necessary. i.e. being able to crop to a small portion of an image of a painting to confirm the condition of the paint and canvas to document its condition prior to shipment when there is apparent damage following the shipment.

With architecture its an ability to capture the finest detail in an image, then being able to hold the detail when preparing the image for final use.

Also, documenting collectables, including automobiles. It would be nice to deliver an image to a client that could be sent to a potential buyer where they can see the entire vehicle but also zoom in to see if there are scratches on the chrome or paint.
 

fotografz

Well-known member
Well - why do you think Sinar and Hasselblad are selling multishot backs if nobody needs them ?

These are mainly used for extremely detailed productshots, reprography of e.g. cultural heritage, paintings and plans.

Architecture Photography for the travelling photographer - remember stitching when not doing a flat stitch will require nodal point operation, usable foreground and many more nuisances.
A Sony A7RII can shoot 5 images a second, so for a 4shot that´s less than a second. a 16 Shot could happen in a bit more than 3 seconds !
Thats also definitely faster than stitching avoiding ghosting of moving subjects (a bit) better than doing it manually (which also holds a source of human failures forgetting a shot etc.pp.)

Pushing such a body into this range of gear would blow the market into a whole new ocean of opportunity.
Sure Blad and Sinar won´t like this, but this is what is going to happe anyway. So I´d preferre sooner, than later.

In short - this would be a killer feature - probably Sony will do it in an Alpha 9R Pro Body, maybe together with USB3 internal and external data bus and a bigger battery ? :)

Listen to me Sony - here you got the lawn and it needs mowing badly.
Give us a mower !
Just to be clear, existing four shot MFD micro step backs do not increase resolution. A Hasselblad 50 meg Multi-Shot back is 50 meg in single shot mode, and 50 meg in four shot mode. The difference is that color separation and moiré suppression is much better with the four micro-step backs that capture each distinct color verses interpolation. This color fidelity also increases the perception of detail.

Older MS backs like those from Imacon and Sinar also offered 8 shot and 16 shot modes which did increase resolution at a time when 16 and 22meg was the highest resolution single shot capability, and there was need for more resolution in certain commercial or institutional situations. Museums particularly needed the accurate color rendering of these backs.

I worked with a Hasselblad 39 meg 4 shot MS back for years, which produced incredible files, matching my Hasselblad H4D/60 in perceptual detail and easily exceeding it in color fidelity. A lot of work I was doing at the time required the color acuity of a MS back.

The current Hasselblad H5D/200 and H5D/200c produces a 50 meg single shot, 50 meg four shot, and a 200 meg six shot file (up to a 400 meg RAW or 600 meg Tiff file!)

MS Backs (as it now works), require hardware and software working in tandem during each shot to capture separate frames ... then proprietary software combines them into a single file, so it is strictly a tethered operation. Generally, it requires a locked down tripod, static subject, and consistent lighting (when working with strobes a user selected delay between each individual shot can be set to allow the strobes to fully recycle). When there are static subjects and those with movement, some photographers would do a single shot then combine it with a multi-shot in post (for example, an interior architectural shot with wind movement outside the windows).

Whether the whole operation can be contained in a camera body including the data combination would be the question. Sony does something similar with rapid bracketing and combining but as far as I know that is strictly a jpeg capture. I suppose four shots could be taken individually and then combined later. The other question would be one of whether that sort of single pixel level precision in movement would be possible in such a small body.

- Marc
 

Stefan Steib

Active member
Using a Multishot Sensorshift on a smaller format like 24x36mm would immediately bring another advantage:
Think about watches or jewelry ? A smaller format can better capture depth of field and using this with a 6 shot, 8 shot, or even 16 shot can achieve an image quality beyond MF.
About tethered: these smaller cameras are MUCH faster than MF backs, the internal buffers can take a full series of shots before they have to be written. Again- superior to MF.

And finally: Pricetag. How much is the H5D/200 now - 35000 $ ? A factor of 10x ?

If that is not an argument, what else ?

Regards
Stefan
 

Annna T

Active member
Using multishots for color fidelity makes more sense IMO. The Olympus E-M5II for instance use four shots for color fidelity avoiding interpolation and then move the sensor by an half pixel both vertically and horizontally, before shooting a second series of four shots. Look at DPreview comparison tool : in highres mode the E-M5II shows absolutely no moire compared to the D810 or the A7r2.

Olympus reps told in interviews that they are working on that feature, with the goal of allowing handheld shooting up to as fast as 1/60 sec. Doesn't mean that they will be able to reach their goal, but interesting to know where their research department is heading.

I think that view cameras have a future if they concentrate on small and light systems able to adapt bodies like the A7r/r2 on the back standard. Arcaswiss has an interesting offer for that. Alas my Arca 6x9 Classic has no geared controls so it isn't worth mounting such a back standard on it. Plus the lenses I have are way too big to allow enough movements, would be too long for the 35mm sensors and I have doubts concerning their performance. I'm not even sure I could mount it on my system.
 
V

Vivek

Guest
I think a focus stacking app like Olympus for macro is more appealing if the Sony 90 G is up to scratch.
 

fotografz

Well-known member
Using a Multishot Sensorshift on a smaller format like 24x36mm would immediately bring another advantage:
Think about watches or jewelry ? A smaller format can better capture depth of field and using this with a 6 shot, 8 shot, or even 16 shot can achieve an image quality beyond MF.
About tethered: these smaller cameras are MUCH faster than MF backs, the internal buffers can take a full series of shots before they have to be written. Again- superior to MF.

And finally: Pricetag. How much is the H5D/200 now - 35000 $ ? A factor of 10x ?

If that is not an argument, what else ?

Regards
Stefan
Don't get me wrong, I'd love to see a FF 135 format Multi-Shot. I've experienced the benefits of M/S, and to have it in a smaller format would be great.

I do tend to agree with Vivek that Focus Stacking would be a more logical solution to control DOF, regardless of format.

Not sure internal "buffer speed" is the key issue with Multi-Shot. Generally (not always), these backs are used in tandem with studio lighting to assure absolute consistency shot-to-shot. When shooting smaller objects using smaller apertures for DOF, a LOT of light is required ... so the timing aspect is more about full recycle speed than how fast the separate images are captured. In many cases the time between each separate shot has to be adjusted longer to accommodate the lights. The MF Multi-Shot backs allow this adjustment to be controlled.

The math doesn't seem to support the notion of a 6, 8, or especially a 16 shot 35mm Multi-Shot camera. The A7R-II does 5 FPS Bursts, so that'd be around 1/4 second per shot ... so any movement in the scene would be recorded and out of whack ... plus, can the sensor be moved that fast? That shows the challenge that Olympus is facing to even get to 1/60 shutter for 4 shot hand-held stuff as mentioned by Annna T.

IMO, the other short-coming of 135 format work of this type is the lack of leaf shutter lenses in multiple focal lengths. Again lighting control is of paramount importance for much work involving Multi-Shot applications, especially involving color fidelity. Higher sync speeds allow the photographer to kill any trace of ambient ... cameras like the A7R/A7R-II are currently stuck at 1/160, and HSS is generally not an option because of the light levels and consistency necessary for this sort of work.

As it stands now, Multi-Shot is a pretty special application with color acuity the main goal. The 39 meg M/S back I used was 4 shot so there was no gain in resolution, but at 39 meg, almost 645 format, resolution wasn't the issue, clarity of color separation was. BTW, it cost nowhere what the H5D/200 does ... which is a VERY specialized tool.

It'll be interesting whether MS gains traction in smaller formats ... personally, I'd love the option, but would much prefer that companies like Sony tidy up their interface before venturing off into less used features.

- Marc
 
Top