rscheffler
Offline
• • • • • •
Upload & Sell: Off
|
 p.1 #9 · AI for advice on choosing lens or camera | |
Scott Stoness wrote:
Last month I went to Churchill to take pictures of polar bears. It was a great trip. https://www.fredmiranda.com/forum/topic/1920108/1#16933750
I fussed over 600/f4, 200-400f4, 200-800, and 100-400 and 1.4x and combinations.
I even went out with 200-800 and 200-400 with 1.4x and 2x to try. This was not overly helpful because of low light and hard to juggle lots of lens and analyse. But it roughly informed me that r7 was a good plan for big animals (bighorn that I found).
At the last moment I was informed that my 600f4 was too big for the airplane.
I got some feedback on this forum which was good but I always worry about bias (and not asking the right questions)vbecause users have the camera and lens they have and biases that arise. So I tried google's Gemini AI - the free version.
After much Gemini-ing I chose to take my r7 (bought it for trip) , r5, 200-400 [with 1.4x (rented it), 200-800 and 100-400 as safety blanket] because:
1) I chose 200-400 for versatility (zoom 200-400f4-560f.6- 896mm/f5.6/r7) Gemini recommended shutter speed f4/f5.6 as more important than 200-800 reach at high fstop. I was initially inclined to 200-800 for size on plane but took the mount off temporarily and put it back on at arrival for flight space. 600f4 was not possible at all for flight.
2) I mostly used r7 for reach because most long shots were cropped and r7 natural cropping on 32mpx was good in good light. This was contrary to Gemini because Gemini was mostly concerned about iso performance and it did not reflect the practical - it's hard to switch and I don't want to crop to 17mpx with r5. So I concluded that r7 in good light was good choice for default.
3) I used efcs for r7 because of slow shutter read speed - Gemini convinced me that otherwise the show read speed would cause warp at long mm because of slow r7 read speed.
4) Gemini convinced me not to use 1.4xexternal on 200-400 for defraction increase related to r7 pixel density that I would not have thought about for really long and added distortion/fstop of 1.4x on top of 1.4x.
4) I used the r5 when it got darker to get my best shots because we were able to get close to mother and cubs late in the day.
I did not use the 200-800 or 100-400 (too much juggling), I mostly used the r7 (for reach), I did not put the 1.4x on my 200-400 because Gemini convinced me that shutter speed and defraction at high fstop would negate 2x effective.
---------------------
Anyway the point of this post is - I found Gemini to be excellent in helping me choose which lens and body and settings to take combined with my practical experience (can't bring it all, can't switch much in field).
I also found Gemini to be very helpful in figuring out how to set up the (new to me r7) camera in menus. When I got stymied by missing option in menu it would quickly give me an answer - much better than the manual. It convinced me not to use electronic at 900mm effective range because of slow r7 read speed. It also convinced me that 2x (higher fstop, more distortion magnified) was not a good option.
It was also helpful in giving very quick informed response, after asking the right questions.
So while I find this forum to be great (friendship and practical), Gemini for the advanced user who knows what questions to ask is great. Gemini is free with flash but it is pretty good. I am sure Chatgtp would do same but when I tried it prior it stopped giving me prompt free advice after some use. It seems to be really good at math problems (fstop, shutter speed, sensor read speed) but not good at realizing the practical - and I think if you were a novice you would not know the right questions to ask to get the right answers. And it gave a fast response with quality that did not need as much judging (is the poster biased or not well experienced).
...Show more →
It's interesting to see the process you went through, the rationale and your post-event summary from actual experience after putting the suggestions to use.
Based on your experiences, it would seem an alternative to the 200-400, if you're not going to go longer than 560/5.6, is the 100-300 with both TCs.
I'm a longtime owner and fanboy of the 200-400 but recent experience with the 100-300 is starting to bring me around to it: it's sharper, higher contrast, handles backlight veiling flare better. Faster AF and takes TCs extremely well on 45MP. Certainly better than the 200-400, which would also translate to the higher pixel density of the R7. Additionally, it's smaller and lighter. A downside is the time needed to swap a TC in/out and to actually have to remove the lens from the camera to do so (not ideal in poor weather or time limited situations).
The rumored 300-600/x.x-5.6 would likely be another candidate, if it existed.
Regarding the R7 settings choices: I would not rule out use of e-shutter in relatively static situations. It's polar bears and not BIF, so rolling shutter would likely be less of a factor. And as I've experienced shooting sports with the R6II (granted, its sensor readout is 2x faster than the R7's), when rolling shutter is present, it's often not noticeable in standalone images (you need to view a sequence of images in rapid succession to notice it). The EFCS tradeoff with the R7, apparently if in H+ advance, is the possibility of shutter shock with certain shutter speeds. Maybe you were consistently outside this danger zone?
What was the airline problem with the 600/4's size? I expect whatever carry solution you had in order to bring both the 200-400 and the 600 would have pushed standard carry-on size restrictions. In the past I've carried on a super-tele bare/loose because the case/bag had to be gate checked. Point being there are likely options, though not necessarily ideal ones. But you also would have then been juggling two large lenses out in the field, which presents its own challenges.
|
) asking good questions.
, the software assembling the final RAW file, I can only guess and likely miss most of it, probably understand even less. It is HERE were most of the physical diffraction gets blurred. I highly doubt that any AI would get to sniff around in this bubble and then fart it out in free interspace.