Abstract: Using the Sony FDR-AX100 consumer camcorder and recording 4k video, 30p frame-rate at 100mbps, under overcast to full sunlight conditions no visible differences were observed - the footage was viewed on a 27 inch 4k monitor - when the shutter speed was varied between 60th up to a 3000th of a second, exposure being kept constant by deploying the built in ND filters as required and gain locked at 0.

Under the conditions of the experiment which included various levels of subject motion - passing motor vehicles, water splashes and ocean waves of various sizes - it appears that the old shutter speed 2x frame rate golden rule does not apply.

John,

Could you give a little more explanation about why you say the "old shutter speed 2x frame rate golden rule" doesn't apply?

I'm not sure to what it doesn't apply. I always thought that it was not a rule, but a guideline. Different shutter speeds are useful for different effects and, on every camera I've ever used including the AX100, the effects are very different.

Phil

He is most likely referring to locking the shutter to correspond twice to the frame rate which in most cases produces less shudder/ shimmering and interline Twitter plus helps SOMEWHAT when panning.

24p. 1/48 shutter

30p. 1/60 shutter

60p. 1/120 shutter

Thanks, Lou., I understand that part, but I was a little unsure about why he said it "does not apply."

Rule or guideline; some say one - others say the other. I'm easy.

The reason I wrote 'does not apply' is because under the various conditions of the test I did not observe any differences when the shutter speed was changed; the range being between 60th and 3000th, all else being kept constant. I did not observe any 'different effects' whatsoever. To use other words shutter speed variations between a 60th and a 3000th did not produce any different effects under the conditions of the test.

I didn't pan or zoom while rolling, I seldom do, other than when following a moving target.

If one does a search on the Internet we learn that shutter speed variations do produce various effects. That is one of the first rule/guidelines I learned many years ago when I first picked up a movie camera. Many still speak of the 180 degrees setting as being optimal and deviating from that will cause the sky to fall in. Not so, at least with my AX100 set up as described.

And this begs the question why not just go with the built in ND filters and lock in shutter speed? The AX100 is capable of shooting very reasonable 20 megapixel photographs. One press on the very conveniently located button above the start/stop record button and one is in photo mode. If, as is usual, I have been shooting 4k 30p video the shutter speed is set to a 60th. Switching to photo mode a shutter speed of a 60th is seldom fast enough and I have been routinely resetting the shutter speed to around a 500th, and removing ND filtration accordingly. While this reset is a bit fiddly the benefits of one cam for video and stills makes this a viable option. During the heat of the battle there were times when I just went ahead and began recording without resetting the shutter speed to a 60th. No harm done.

I have always looked for new gear that has built in ND filters. While I also have a set of screw on ND's and various adapters when in the field setting that up is a bit of a pain therefore built in ND's was high on my list of required features.

I asked a fellow who tests many cams, especially smaller models where built in ND's are rarely possible, about his thoughts regarding using shutter speed to control exposure. He does it all the time with no ill effects, he told me (not his exact words).

Just to be clear I am not saying different effects can not be generated by varying shutter speed. I am saying that under the conditions of the test, which is how I routinely use the AX100, shutter speed variations did not produce a visible difference.

Built in ND filters is about to be removed from my list of important features, on a case-by-case approach, I should add.

"If one does a search on the Internet" one can learn all sorts of things, not necessarily correct. The 2x frame rate (180 degree) shutter rule is a reasonable rule of thumb - it was never golden. If you shoot with a shutter around 180 degrees, say between 145 and 270, motion blur from action and pans matches most movie film. Over 145 with action you're likely to notice strobing, especially with 24-25-30p video. Doesn't seem to bother a lot of TV footage. Good on you John for doing some experimenting and passing on your results rather than blindly accepting convention.

Dang - accidentally double posted earlier edit to remove "spurious". John, very sorry - didn't mean to suggest your results were in any way false or misleading, just worthy of further study

Spurious is a big word (rhetorically speaking) therefore I checked Merriam-Webster. We are informed '...the more common meaning is "false" (a sense introduced to spurious in Late Latin).'

Here again, in a nutshell, let me repeat, is the result of the test I conducted:

'...under the conditions of the test...shutter speed variations did not produce a visible difference.'

In what way is the result 'a bit spurious' (false)?

Please understand my reference to 180 degrees was to historically contextualize the test. Please also pardon me for the possible confusion; this test is not about traditional film (celluloid) movie cameras but the Sony FDR-AX100 digital 4k capable consumer camcorder. Shutter degrees is not mentioned in the AX100 online help guide.

I do appreciate your thanks for sharing the results and while I am tempted to bask in your mild praise for not blindly accepting convention I must in all honesty emphasize that was not the driver. As described above this test was driven by error and then pragmatism, that's all. Having said that I do subscribe, tongue in cheek that is, to the maxim attributed to the Postmodernists 'question authority and ignore the answers', but that's a bit off topic :-).

Shooting stills and video thru the same lens at the same time has been at least for me a holy grail quest and I've tried pulling it off over the years with an HC1, FX7, cx550v, a5100 and ax100
During a recent figureskating competition and using ax100 indoors arena with full spectrum florescent overhead lighting which provide far fewer lumens than sunshine, in fact 12-15 db of gain is often required and using f5.6 - 8 in order to achieve a decent dof to keep the fast moving figureskater in focus it is then a tricky matter to use shuttter speeds faster than 1/60. Of course the shutter needs to be 1/250 to stop motion but there is insufficent light so the iso goes up and the images get grainy. There is also a perceptual loss of colour depth or something which is hard to describe plus and this sounds weird but there's almost a cumulative effect where the video catches up to the frames
as always plenty of light is better than slightly insufficient light.
I find that the ax100 takes forever to write a photo file and doubly long to write a photo file while recording video so extracting 4k stills works for me but does take some time after the fact

This was discussed at least peripherally in the never ending AX100 thread when the cam first came out.

I initially let the camera run "auto", including shutter speed - the results were horrible, with strobing or "shimmer" on all detail with any motion at all. The "effect" was that of a series of very sharp stills (sort of like a "flip book"), and it was jarring....

Once I started riding the shutter speed and keeping it closer to the FR, I got much more pleasing results, as there was now motion blur to help smooth out the overly sharp "stills". I fiddled with up to around 125 shutter with still decent results, so the "rule" is bendable to SOME degree, depending on one's tolerance for the video becoming "jumpy" or strobing/shimmering.

Of course if one is pulling stills from 4K clips, or shooting stills, motion blur may not be desirable (and higher/faster shutter speeds ARE), so it becomes a balancing act depending on the motion and detail in the scene....

Funny how shooting the latest greatest 4K requires one to go back to the old school fundamentals of photography and cinema shooting guidelines! I actually enjoy shooting more though, as I have to "think" a bit more about what I'm wanting the camera to do, rather than relying on the camera being "smarter" than I am.... it's probably still smarter, but at least I get to tell it what to do to get closer to what I want!

I get the impression that you and I, Bruce, are not the only ones always on the lookout for the perfect photo and video cam and while the AX100 is the best I have found to date it is far from perfect. For serious wildlife action, like a couple of days ago when I encountered a flock of Australiasian Gannets feeding, diving from 20 to 30 meters, then the AX100 just doesn't cut it and I reach for the Sony A6000 with a Sony FE 70-200mm F4 G OSS lens attached. That certainly did the trick.

I also find the photo mode on the AX100 slowish however when the amount of gear I can bring along is limited and the daylight is plentiful, as is usual here, then the AX100 does a half-decent job, especially now that I don't need to fiddle so much with the controls.

Yesterday, high cloud with a weak somewhat insipid sun trying to do its thing, I shot a bunch of clips with the ND filter off, gain locked in at zero, using the shutter to obtain correct exposure with speeds from 200 to 1000th and again absolutely no suggestion of shimmering or interline twitter or any other aberration for that matter in any of the clips.

Dave, using my AX100 under the conditions described above the rule is not just bendable; it is broken. To use less dramatic language let me say the 'rule' is N/A (non-applicable - does not apply).

I have no idea what I am doing 'wrong' to get such excellent results at such outrageously high shutter speeds (now including a 4000th). Maybe a firmware update or two, but there is no getting around it; what I am seeing on my 4k monitor is nothing but gorgeous looking-out-an-open-window seascapes and landscapes we know the AX100 is capable of delivering. And not the faintest suggestion of artifacts of any kind.

That'll do me!

daylight and lots of it is helping your shooting senario forsure

All my shoots are in the theatre. All cameras set at 1/60 shooting 60P on the AX100, AX53, NX30U and 60i of course on my now aging NX5U. For shows I even take stills, in post, from any of the 60P cameras and they are not bad !! I think the real issue with the "rule" was for slow frame/sample rates as any camera movement could cause artifacts like judder. Once you move to 60P camera movement is less of an issue and mainly goes away.One can then choose shutter speed for artistic reasons. Do you want blur or pin sharp image. If it is too sharp and high resolution one may need to go to an even higher frame rate too. We shall see what happens with 8K and HDR I expect a higher frame rate.

Ron Evans

There's a definitel interaction between frame rate, shutter speed, and so on, along with the patterns and action in what you're shooting. Plus whether you want motion blur or super sharp individual frames. Lo resolution won't "cover up" any mistakes when you go to 4K!

I'd prefer to have 4K/60p, but for now 30p is what is available at an economic price both for format and for memory cards. It probably doesn't help that my big screen 4K display/playback is also 30Hz... but again, it is what it is for the price, and it still looks stunning.

If higher shutter speeds are working OK for you, that's good, I was just taken off guard at the shimmer I saw when I let the camera run the shutter speeds up. Now I'd push the ND filter into play to try to keep the shutter speed down, it's convenient with the AX100! The early problems were pretty much "out of the box" before I had time to figure out the camera, once past the learning curve it's been an outstanding camera that I don't think you can beat for the $$.


I'm sure you're aware of the RX10M3... aside from probably having to get external ND filters (dumb thing to leave that built in ND feature off, Sony!!), that may turn out to be the "best" dual mode camera... I've already been using the RX10 and the RX10M2 for my video purposes.... Still have the AX100 for what it does well, but the RX has been good to me... For multicam events, the AX100 and RX's are complementary for the most part, and all amazing cameras image wise.

.

Ron, I assume you are shooting HD with the AX100 in order to obtain 60p. I do recall your serious preference for higher frame rates. I too subscribe to the notion of faster-is-better frame rates and that's one of the reasons I imported an NTSC model AX100 maximum 4k frame rate of 30p vs. purchasing locally a (less expensive) PAL model maximum 4k frame rate 25p.

My next purchase might well be a Handycam that shoots 4k at 60p. That would be nice. While the popular 1 inch sensor as deployed in the AX100, and more and more cams as time passes, I suspect (shall we call it) the AX200 which delivers 4k at 60p is not just around the corner. I would be extremely happy to be wrong of course, but it seems that the smallish Handycam form factor, which I enjoy, presents serious challenges to the manufacturer not the least of which is heat dissipation. I would also like a brighter EVF than on the AX100.

Whatever, the 4k capable AX100 with the shutter speed flexibility, even at 30p, is certainly a hard act to follow. I'm not holding my breath waiting on 8k, HDR and who knows what higher frame rates :-)

Thanks Dave, and yes, I certainly agree that the ND's are a convenient and efficient way of managing exposure. It was only the switching between photo and movie mode that got me checking out shutter speed effects, and the apparent lack thereof.

Yes, I am aware of the RX10 series. I have the original and before I acquired the AX100 it was my go to cam almost invariably. Now it sits forgotten somewhere or other. I too was somewhat surprised Sony did not include ND's on the RX10 III to the point that I decided against going with that model (one of the reasons). I presumed it was because of the already humongous lens on the cam and adding built in NDs would have been just over the top altogether.

Regarding the ND-less RX10 III perhaps, as with my experience with the AX100 that is the subject of this thread, high shutter speeds may not cause problems. Have you tried simply cranking up the shutter speed to reduce exposure in lieu of ND's? That might be interesting...

John, I have had the FDR-AX1 since they came out and in good light it produces a nice image at 60P. I am going to look at the Panasonic AG-UX180 if/when it appears in the fall. That has a 1" sensor , shoots UHD 60P, 20x zoom and hopefully a nice match for the other cameras I have but will be a replacement for the FDR-AX1 and maybe even my NX5U. Probably has a Sony sensor !!! If Sony had put UHD 60P on the PXW-Z150 it would have been the camera to get. I am sure they could have and choose not to for some strange marketing reason !!! I am an all Sony guy at the moment but this fall may be the start of a change,


I too like the AX100 even though I shoot 1920x1080 60P exclusively on it.


Ron Evans

@Ron - the "HD" from the 1" class sensors is pretty good, but once I got the shutter speed tamed, 4K has been my preference (even if I wish it was 60P).


@John, I think it will be a while before we see 4K/60p, as it's really at the bleeding edge for the entire chain from acquisition to display. Maybe 3-5 years... probably more? And unless you're dealing with absolutely HUGE screens, these other higher resolutions are probably a bit of overkill (even well shot HD looks good on a 4K screen).

I've got the RX10 sitting around as well, I've been saving it for a couple still shooters I know that are
"thinking" about an upgrade, and it is a darn fine camera for stills and HD video, any way you slice it! Eventually, it may go to ebay...

i grabbed a "Mk2", and that's my main/secondary cam alongside the AX100 right now. The M3 is on my "hit list" once the prices drop a bit on a "lightly used" one.. I'll decide on the ND once I've used it a bit, I'd much prefer that it's in the body like on ALL the other RX's (10 and 100), but that new long lens means it will replace multiple lenses and an aging small chip superzoom, so once that stuff is sold off, I can afford a decent variable ND <wink>.

The M3 is expensive, bigger and heavier, but for me that lens is a huge attraction! iF I have to kluge the filter setup, so be it... the lens is a bit slower than the f2.8, and so maybe I can just stop down to slow the shutter speed.... I think the M2 will probably be sticking around as a backup anyway if I need the ND feature!


The Sony Cybershot team seems to be a bit more aggressive with product releases, even while retaining the basic design elements (for the most part). I'm on the MkIV of the RX100, they keep dribbling upgrades (sensor primarily, and at least a rudimentary 4K capability (due to the heat issues in a teeny, tiny body). The RX10M2 doesn't have heat issues, has the updated sensor, and was an easy choice... the M3 was a surprise, but makes sense to me. I'm intrigued to see what they might cook up next!

Logically, there SHOULD be a AX100M2 with the upgraded sensor, yet the Handycam division seems intent on pushing the small chip/BOSS format (now on it's second iteration). I fear that the AX100 may not EVER be dupiicated/updated (although I feel it is a "reincarnation" of the venerable HC1 in many ways). It would be a shame, but I'm OK with not feeling there is any compelling reason to "upgrade" the AX100 anytime soon!

I've wondered if at some point in time Sony will integrate or merge or whatever the "consumer" Cybershot and Handycam lines to improve efficiency - the low end video and P&S markets are "mostly dead", so pretty much anything underneath the "enthusiast" stuff is "dead money" for R&D or market share (a shrinking fraction of a vanishing market is NOT where Sony will put it's money). Since they just sold off Vegas, so restructuring wouldn't be a surprise...

@John, I think it will be a while before we see 4K/60p, as it's really at the bleeding edge for the entire chain from acquisition to display. Maybe 3-5 years... probably more? And unless you're dealing with absolutely HUGE screens, these other higher resolutions are probably a bit of overkill (even well shot HD looks good on a 4K screen).

...

Most of the 4K TV's in local stores are now 60P with HDMI 2.0. Panasonic this fall , if they are on schedule, will have the AG-UX180, what I had hoped the PXW-Z150 would be , 1" sensor, 20xzoom, UHD 60P to SDXC U3 cards with SDI and HDMI 2.0 for under $4000. Unless you are a real Sony fan or need streaming the PXW-Z150 will not be competitive. Probably uses a Sony sensor too. I use EDIUS 8.2 to edit that has no problems editing UHD. So I do not see the transition to be a long time away. I have had my FDR-AX1 that does UHD 60P for almost 3 years and yes that is sort of bleeding edge since it took a long set of updates to not quite finish the product. I think the real problem for 4K TV's is that SD input looks horrible so watching legacy stuff will be a bad experience !!!!

Ron Evans

sd looks just fine on a 4k screen provided the viewing window is sized in proper proportion

Yes I understand, but what consumer is going to do that when playing an old SD DVD from their DVD player to a 4K TV ? And how many TV will allow you to just use a small portion of the screen to view in that way. Not done any research on that but have seen a simple playback of SD on a 4K TV and it is not good. The upscalling to UHD from SD is not good.

Ron Evans

The Panasonic AG-UX180 looks rather much like it shall be my next cam acquisition from the little I know about it and assuming there are no devils in the details. Due October 2016 I understand. In the meantime the AX100 is doing the job quite adequately and for some situations will continue to enjoy a spot near the front of my gear bag.

But a 1 inch sensor, 4k 60p, not costing an arm and a leg, is where I want to be, ASAP.

Dave, using my AX100 under the conditions described above the rule is not just bendable; it is broken

Me again, still embarrassed from my earlier post, but I think this is significant. I always understood the "rule" wasn't related to any particular camera and had nothing to do with electronic technology - in any case a video camera is going to use less buffer at higher speeds. Its all about physiology - do we think footage looks better with motion blur? Another chink in the "rule" came with John's reference to gannets, which probably dive at 100mph and thus still should have motion blur at 1/200th. A while ago I was shooting next to a pro who was happily shooting 50i footage for TV of some boat races at 1/500th, unaware of any such rule. Anyway, I grabbed my nearest camera (a point and shoot PanyTZ70) and panned some veranda slats, handheld, but good potential for motion blur or stobing. I'm having difficulty seeing the difference between 1/50th and 1/250th, and maybe the 1/250th looks better. I've posted the footage on Vimeo. Shutter Speed 2X framerate "rule" on Vimeo I'll experiment further, but I'm inclined to agree - the "rule" is not just bendable, it's broken.

sd looks just fine on a 4k screen provided the viewing window is sized in proper proportion

SD doesn't even look good on a 1080p display of larger sizes (65"+).

No worries, Rainer. While I discovered deviating from the old 2x frame rate for shutter speed guide/rule with my AX100 did not bring about the end of the world as we know it I would hesitate to elevate this finding to a new statement of fact applicable at all times and in all circumstances. I did choose my words carefully pointing out 'under the conditions of the experiment'.

Furthermore, I confess to having no idea what is going on under the hood (nor do I care that much actually).

Let me add a caution: I would not go so far as to suggest this approach as a general new rule. The take home I suggest is that if the cam one is shooting with does not have ND filters and you don't want to fiddle with screw-ons then try upping the shutter speed, even as high as a 4000th, to control exposure, thereby seriously deviating from the old guide/rule. But test before shooting something important, as you are doing :-)