Reilly Diefenbach
Forum Pro
If only you knew how irritating that is and were able to adjust your typing accordingly.
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
A7RV too noisy for me
- Thread starter djmParkCo
- Start date
I think the entire point of all these posts are to be irritating?
Hawkypuck
Senior Member
I wish him the best:-D
Hawkypuck
Senior Member
True!
Harder to remove noise! That is why i turn off NR on my camera!
Harder to remove noise! That is why i turn off NR on my camera!
dazzerfong
Senior Member
Hawkypuck
Senior Member
Interesting, on my 7d mark ii it allows me to...
MediaArchivist
Veteran Member
Be not one live to see, inside is seen.
gil
Forum Pro
OP was supposed to be a professional wildlife photographer with ads for photo tours. Strange that he was so particular with that boring BIF - with halos, subject at undesirable location (of course could be cropped from that 61MP) and a so-so subject. Why that pic? It does not represent him being a pro wildlife photographer. Of course the exif showed a higher ISO than what normally would be used at that wing spread and with seemed to be adequate light (goose belly was lighter).
How many pics were taken at that condition/similar exif? Were all pics showing that much noise at 100% look? I don't own nor have used a 61MP camera but my A7IV also showed some similar noisy BIFs against overwhelming acceptable pics on the same session and condition. One or few noisy undesirable BIF pic (for whatever reason) against 900 or more BIF pics with better subjects (and BIF wing spread) is a non-issue to me but of course I am not a pro and with a different name
.
--
Cheers,
gil - San Jose, CA
95% Cheap Lens,90% JPG and 100% Handholding Provocateur
Like happiness, photography is often better created than pursued.
How many pics were taken at that condition/similar exif? Were all pics showing that much noise at 100% look? I don't own nor have used a 61MP camera but my A7IV also showed some similar noisy BIFs against overwhelming acceptable pics on the same session and condition. One or few noisy undesirable BIF pic (for whatever reason) against 900 or more BIF pics with better subjects (and BIF wing spread) is a non-issue to me but of course I am not a pro and with a different name
.--
Cheers,
gil - San Jose, CA
95% Cheap Lens,90% JPG and 100% Handholding Provocateur
Like happiness, photography is often better created than pursued.
Last edited:
abe4652
Senior Member
I recently shot (not for money) an entire engagement for a friend's daughter just to give him some photos and accidentally had it set at ISO 8000 on my A7R4. My fault, but I only had 45 minutes notice and didn't check prior settings.
A quick trip though DXO in Deep Prime (not the latest further Prime) and all was more than fine. A portrait at 61mp will look more than good enough when run through DXO, and there are no abnormalities from having gone through noise reduction.
Hi,
I saw Don's tips but I am not sure how his video linked to your point.
I echo the few post I saw about high mp does not = more noise. I suggest you take a look at Tony's video. If you agree, it does help to update your first post so as to avoid confusion for future reader.
I saw Don's tips but I am not sure how his video linked to your point.
I echo the few post I saw about high mp does not = more noise. I suggest you take a look at Tony's video. If you agree, it does help to update your first post so as to avoid confusion for future reader.
Pakorn KPSD
New member
- Messages
- 1
- Reaction score
- 0
For me to reduce Noise for Sony A7RV
Sharpness setting:does actually control the amount of Noise.
Creative Look setting I set Clarity to 0,and play around with Sharpness and Sharpness Range depend on my subject.
Sharpness setting:does actually control the amount of Noise.
Creative Look setting I set Clarity to 0,and play around with Sharpness and Sharpness Range depend on my subject.
blue_skies
Forum Pro
You crop at 100%, or better, pixel peep, and then you find out that higher res sensors have more noise?Recently bought an A7RV and already returned it for a refund. Nobody seems to be talking about the super high noise levels of this camera. I own and currently us an a6600 and A9II and both those cameras are WAY better regarding noise. Of course those both have 24PM sensors and when you jam 61MP onto the same size sensor (A9II) of course the noise level will go up. I knew that and I also know that current PP techniques can manage the noise well. No thank you. I figured maybe the newest processing power might take care of some of the noise - nope. Even at moderately reasonable ISO's such as 500 or 1000 the noise on the A7RV is crazy high. In case you're tempted to tell me it's operator error look here:
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000 (100% crop of above)
After which conversion? OOC, RAW, Editor (choice)?
Denoising works quite well, even with the higher res sensors.
But, a basic rule: noise and sharpening should be done with the desired final resolution in mind. If that is 24Mp, then you'll find that the 61Mp produces identical results.
Of course, at the 100% crop detail, the 24Mp would look quite poor - lack of detail.
It doesn't require to purchase, and then return, a camera body to realize your mistake...
--
Cheers,
Henry
Entropy512
Veteran Member
https://www.photonstophotos.net/Charts/PDR.htm#Sony ILCE-6600,Sony ILCE-7RM5,Sony ILCE-9M2
The R5 actually outperforms the A9II significantly at low ISO.
Edited to add the A6600, which gets (as expected) utterly crushed by its FF brethren. (Except for the A9, where it's clear that image quality was traded for speed at low ISO)
Edit 2: Also of importance here:
https://blog.kasson.com/the-last-word/noise-reduction-with-nonlinear-tools-and-downsampling/ - higher-resolution cameras are superior if PDR is similar.
The R5 actually outperforms the A9II significantly at low ISO.
Edited to add the A6600, which gets (as expected) utterly crushed by its FF brethren. (Except for the A9, where it's clear that image quality was traded for speed at low ISO)
Edit 2: Also of importance here:
https://blog.kasson.com/the-last-word/noise-reduction-with-nonlinear-tools-and-downsampling/ - higher-resolution cameras are superior if PDR is similar.
Last edited:
Entropy512
Veteran Member
Actually better with higher-res sensors - https://blog.kasson.com/the-last-word/noise-reduction-with-nonlinear-tools-and-downsampling/You crop at 100%, or better, pixel peep, and then you find out that higher res sensors have more noise?Recently bought an A7RV and already returned it for a refund. Nobody seems to be talking about the super high noise levels of this camera. I own and currently us an a6600 and A9II and both those cameras are WAY better regarding noise. Of course those both have 24PM sensors and when you jam 61MP onto the same size sensor (A9II) of course the noise level will go up. I knew that and I also know that current PP techniques can manage the noise well. No thank you. I figured maybe the newest processing power might take care of some of the noise - nope. Even at moderately reasonable ISO's such as 500 or 1000 the noise on the A7RV is crazy high. In case you're tempted to tell me it's operator error look here:
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000 (100% crop of above)
After which conversion? OOC, RAW, Editor (choice)?
Denoising works quite well, even with the higher res sensors.
I disagree with this - noise reduction should be done as early as possible in the pipeline (before downsampling). Also downsampling should be done on linear data (Some of DPR's comparisons unfairly penalize high-res sensors in the studio comparison tool, I suspect due to nonoptimal downsampling strategy). Unfortunately I can't find one of the good reference links describing why downsampling in nonlinear space is a bad idea at the moment.
--
Context is key. If I have quoted someone else's post when replying, please do not reply to something I say without reading text that I have quoted, and understanding the reason the quote function exists.
Paul JN
Leading Member
Actually the SOOC jpegs are very good at high ISO. I did some comparisons at high ISO and was very impressed with the jpeg lack of noise. The chroma noise was better than what LR could do on the RAWsOne of his complaints was that the camera itself was not doing the denoising when saving the RAW images. In fact there are cameras that do that (all the recent Canon cameras starting from the R5), so you think it's good?Given these examples, one could argue that the a7R V has better control over its noise since it has more pixels than the A9II.I see you used a 100% crop image above, so let me guess: you compared 100% crops from the A7RV with 100% crops from your A9II? Big mistake and not a valid comparison.Recently bought an A7RV and already returned it for a refund. Nobody seems to be talking about the super high noise levels of this camera. I own and currently us an a6600 and A9II and both those cameras are WAY better regarding noise. Of course those both have 24PM sensors and when you jam 61MP onto the same size sensor (A9II) of course the noise level will go up. I knew that and I also know that current PP techniques can manage the noise well. No thank you. I figured maybe the newest processing power might take care of some of the noise - nope. Even at moderately reasonable ISO's such as 500 or 1000 the noise on the A7RV is crazy high. In case you're tempted to tell me it's operator error look here:
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000 (100% crop of above)
Here I made a valid comparison for you. I took the ISO 1600 raw files from the DPR study test scene of the same two cameras, A7RV and A9II. Used RawDigger to create an RGB Render TIFF of each file, ie free from any software noise reduction and sharpening. Normalised the white balance and exposure of both files, and enlarged the A9II file to same size as the A7RV.
Now look at the neutral grey squares on the test card section of the DPR studio scene:-
A7RV top, A9II bottom. ISO 1600
No difference worth mentioning.
Clear?
And I won't bother doing it for the a6600, because it will be worse.
cheers
I thought the same thing as djmParkCo way back in 2008 when I got my Canon 21 MP 5D2 and compared it in similar fashion to my Canon 12 MP 5D. "Oh my god, look at that noise" in the sky. So, I understand the OP. Hopefully, djmParkCo sees his way back to the a5R V if he is a professional BIFer (or the A1).
The A7siii is doing that too, applying DNR on the video file itself.
The thing is that most photographers don't want that post-processing step to be done in cameras. Sure, it reduces the noise, but by doing that, it also permanently "freezes" the noise onto the image, and it becomes almost impossible for a high quality denoiser to recover the details and eliminate that noise.
However, when not processed at all, the noise becomes much easier to detect and deal with by those high quality denoisers, and you ultimately get an image with much greater detail and much reduced noise.
=> That is the reason why people don't want DNR to be applied on their output files. However if like the OP, you don't want to do any kind of post-processing, then yes you actually want to enable the DNR (not sure that he did, there is such an option) and you also want to shoot JPEG (again, not sure if the did that).
--
P
ProDude
Guest
I agree. To me LR is useless. DXO Photolab 6 does a fine job opening a RAW file properly adjusting for any distortion and all the rest of the file profile. LR doesn't do that to begin with properly. Their NR is at this time the best there is. Next would be Topaz. LR about dead last.Actually the SOOC jpegs are very good at high ISO. I did some comparisons at high ISO and was very impressed with the jpeg lack of noise. The chroma noise was better than what LR could do on the RAWs
MediaArchivist
Veteran Member
Year after year, there continue to be posts that insist this is not true. Amazing.
blue_skies
Forum Pro
Downsampling itself is a form of noise reduction. In most cases, I am guessing that the downsampled result would be fairly similar to a result obtained with a lower resolution sensor, aside for the linear data effects - I agree with your comments.Actually better with higher-res sensors - https://blog.kasson.com/the-last-word/noise-reduction-with-nonlinear-tools-and-downsampling/You crop at 100%, or better, pixel peep, and then you find out that higher res sensors have more noise?Recently bought an A7RV and already returned it for a refund. Nobody seems to be talking about the super high noise levels of this camera. I own and currently us an a6600 and A9II and both those cameras are WAY better regarding noise. Of course those both have 24PM sensors and when you jam 61MP onto the same size sensor (A9II) of course the noise level will go up. I knew that and I also know that current PP techniques can manage the noise well. No thank you. I figured maybe the newest processing power might take care of some of the noise - nope. Even at moderately reasonable ISO's such as 500 or 1000 the noise on the A7RV is crazy high. In case you're tempted to tell me it's operator error look here:
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000
A7RV 200-600mm lens at 600mm 1/2500 sec. @ f/6.3 ISO 1000 (100% crop of above)
After which conversion? OOC, RAW, Editor (choice)?
Denoising works quite well, even with the higher res sensors.
I disagree with this - noise reduction should be done as early as possible in the pipeline (before downsampling). Also downsampling should be done on linear data (Some of DPR's comparisons unfairly penalize high-res sensors in the studio comparison tool, I suspect due to nonoptimal downsampling strategy). Unfortunately I can't find one of the good reference links describing why downsampling in nonlinear space is a bad idea at the moment.
Lowering noise reduction first makes the images also more pleasing to work with, I understand that too.
Also, any other processing will make the noise reduction process more difficult - e.g., sharpening and clarity sliders might defeat noise reduction algorithms. So, in this context, also noise reduction should be first.
Otoh, I did notice that sharpening after downsampling leads to images that present better at the reduced resolution, especially if the gap is large (e.g., web images).
Imagine anti-alias effects, which could wash out if you downsample after applying them.
I am talking about RAW editors, in JPG processing, things are slightly different.
--
Cheers,
Henry
I’m two weeks in with my RV and I haven’t noticed this.
Of course I run all images through DXO, just because I’m OCD about IQ and it’s so easy.
I understand wanting to shoot JPEG, but I personally wouldn’t buy an almost $4k camera to shoot JPEG.
Of course I run all images through DXO, just because I’m OCD about IQ and it’s so easy.
I understand wanting to shoot JPEG, but I personally wouldn’t buy an almost $4k camera to shoot JPEG.
P
ProDude
Guest
Indeed the profiles from DXO for the A7RV are superb no question about it, far superior to anything Adobe to say the least. But I must say I'm VERY impressed with the out of camera Jpegs the A7RV produces. I've never seen such pleasing results in Jpeg from any camera prior, and considering that covers top Canon, Fuji and Olympus that's saying something. There are times Jpeg does come in handy, I'll just leave it at that.
Similar threads
Keyboard shortcuts
- f
- Forum