VR lens value on the new breed of DSLRS

Kholi said:
Squig man you may wanna take abreak from the forums. LoL

That's a G1,not a GH-1. Don't care how many numbers or letters they share they are not the same cameras, so going by this or that from a previous model doesn't do anything good for you.

Write some more, man. Honestly, if most of us put as much effort into writing as we do into posting we could finish several feature length scripts in the span of just months.

The gear is important, but the script is where it starts and without one (unless you just shoot off the cuff, but then you'd be out shooting though, right?) you dont' get to use the gear.

GH-1 will be another awesome tool with more expanded image controls than those before it.

This topic isn't about that, though, it's about VR Lens validity and purpose.
Click to expand...

yes mum

I'm just a bit anxious about getting the right tool to realise my vision. The technology is there but it's just not being packaged right, it's real frustrating.

I'm writing, I'm writing

oh yeah VR lenses...I have the flu I'm a bit out of it. far too slow.....that goes for the lenses too.
 
The reason why we don't see faster optical image stabilization is because there is an optical element that is moving within the lens to stabilize the image. The size of this moving optical element is the limiting factor for how 'fast' an OIS lens can be. The f-number, aperture, size is determined a ratio the focal length to effective aperture diameter. So the larger the aperture, the larger the moving optical element has to be for the OIS lens.

Another thing is effect of adding this optical element is that there is a marginal reduction in quality since light has to pass through another glass element, which is the reason why Nikon doesn't build VR into their recent 14-24mm F/2.8G or their 24-70mm F/2.8G professional lenses, this is combined with the fact OIS works better with longer focal ranges then short and the IQ cost-benefit of having an OIS for those lenses doesn't makes little sense.

Now, what we may see is a sensor-based IS from an Olympus micro-four thirds camera or whenever video is available for Sony dSLRs. In general, OIS is preferable to sensor-based IS from a performance perspective, however, a sensor-based IS doesn't have the downsides similar to OIS and can be used with any lens, including the ultra-fast primes.
 
very true and i agree the size of the moving element is probably a major reason, but i'd just like to mention that in modern lens designs the optical aperture size and the physical one are rarely the same. you won't find many 200mm f/2's that have four inch elements, right? it's certainly theoretically possible to design a vr lens with a small moving element that's still fast. i think market price is also a factor, fast lenses these days are already extremely expensive, and with vr it becomes ridiculous.

/matt
 
Steve Castle said:
The reason why we don't see faster optical image stabilization is because there is an optical element that is moving within the lens to stabilize the image. The size of this moving optical element is the limiting factor for how 'fast' an OIS lens can be. The f-number, aperture, size is determined a ratio the focal length to effective aperture diameter. So the larger the aperture, the larger the moving optical element has to be for the OIS lens.

Another thing is effect of adding this optical element is that there is a marginal reduction in quality since light has to pass through another glass element, which is the reason why Nikon doesn't build VR into their recent 14-24mm F/2.8G or their 24-70mm F/2.8G professional lenses, this is combined with the fact OIS works better with longer focal ranges then short and the IQ cost-benefit of having an OIS for those lenses doesn't makes little sense.

Now, what we may see is a sensor-based IS from an Olympus micro-four thirds camera or whenever video is available for Sony dSLRs. In general, OIS is preferable to sensor-based IS from a performance perspective, however, a sensor-based IS doesn't have the downsides similar to OIS and can be used with any lens, including the ultra-fast primes.
Click to expand...

Wow. Thanks for that breakdown, Steve. I like the sound of a sensor-based image stabilization system. I can't imagine the sort of damage you could do with a fast-ass 50/1.4 and OIS enacted.

It may be some time before we see that, but the good thing is that the zooms might be pretty adequate for most jobs barring you can control the ISO as well.
 
mattsand said:
very true and i agree the size of the moving element is probably a major reason, but i'd just like to mention that in modern lens designs the optical aperture size and the physical one are rarely the same. you won't find many 200mm f/2's that have four inch elements, right? it's certainly theoretically possible to design a vr lens with a small moving element that's still fast. i think market price is also a factor, fast lenses these days are already extremely expensive, and with vr it becomes ridiculous.

/matt
Click to expand...

You are right, not every lens element must be equal to the maximum aperture, as light moves through the lens light is focused down to the size of the light sensitive area(sensor/film). However, there are lens elements within the Nikkor 200mm F/2 that are ~four-inches/100mm in diameter(its a huge lens after-all), but not the moving OIS element.

In a lens, every element and group is dependent on the preceding and proceeding piece of glass. The larger the maximum aperture, the larger the overall bulk of each lens component (and a larger OIS element). Accomoding a larger OIS element also means that a proportionally large increase in motor capacity and overall capability of VR needs to be attained as well.

Another thing with OIS is that there is an impact of any moving optical element to distortion and overall optical performance. Applied carelessly, image quality may be affected negatively. Hence, lens makers are very careful of where to apply this technology, and is likely the reason why we haven't seen popular implementation in wider primes.
 
Kholi said:
Wow. Thanks for that breakdown, Steve. I like the sound of a sensor-based image stabilization system. I can't imagine the sort of damage you could do with a fast-ass 50/1.4 and OIS enacted.

It may be some time before we see that, but the good thing is that the zooms might be pretty adequate for most jobs barring you can control the ISO as well.
Click to expand...

I'm thinking we may see a sensor-based image stabilization in a micro-four thirds cameras from Olympus. My assumption was that sensor-IS wasn't added to the G1/GH-1 sensor because the flange distance between the sensor and lens was small making it difficult to package it within the form-factor. However, Panasonic has historically been pushing OIS like in their DMC-L1 so that may not be the case.

But sooner or later, whenever Pentax, Sony, etc that have sensor-based IS start offering video it should be possible to use wider-primes & IS. Hitherto, shorter focal-distances have seen minimal improvement using IS, but there have been some improvements recently. The newer Sony sensor-based IS have claimed 2.5-3.5 stops improvement, which is impressive for a non-optical system. And Fuji has shown a dual-system in their bridge camera that uses both optical and sensor-based IS. So there is still a lot of possibilities with the technology.
 
You must log in or register to reply here.
Back
Top