A recent post on another forum made me check my camera (A7R IV) and it seems that I can only get upto 7.5-8fps in Full frame Hi+ node, no matter the setting (e.g. small jpg with manual AE/AF and V90 card). If I switch to APS-C mode, then I can easily get 9.5-10 fps. I checked another A7R IV and found it to do the same.
Has anyone else actually checked and verified that they can get 10fps on their camera? If so, what were the settings?
I recall uncompressed mode slows things down. Not sure if this the reason. I haven’t done scientific measurement on whether mine could actually shoot 10FPS. Did you set shutter release as priority? If not, you may not get the 10FPS if the camera thinks the subject is not in focus.
Maybe 10fps is only possible with the electronic shutter? Check the EFCS settings too.
How are you measuring the speed? I suggest you take the first and last shot of a long burst, and then inspect the EXIF: the exact time will be: DateTimeOriginal + SubsecTimeOriginal / 1000.
DateTimeOriginal only gives the time in seconds.
SubsecTimeOriginal gives the milliseconds.
indusphoto wrote:
A recent post on another forum made me check my camera (A7R IV) and it seems that I can only get upto 7.5-8fps in Full frame Hi+ node, no matter the setting (e.g. small jpg with manual AE/AF and V90 card). If I switch to APS-C mode, then I can easily get 9.5-10 fps. I checked another A7R IV and found it to do the same.
Has anyone else actually checked and verified that they can get 10fps on their camera? If so, what were the settings?
"In Continuous Hi+ mode, the Sony A7R IV's top burst speed is a claimed 10 frames per second, and that's with continuous AF/AE. In the lab, the A7R IV essentially met that spec with JPEGs and compressed RAW files, but the speed dropped to about 7.2 fps with uncompressed RAW files. In Continuous Hi mode which updates the viewfinder image between captured frames for easier subject tracking, the Sony A7R IV met its spec at about 8.0 frames per second with JPEGs and compressed RAW, but it also slowed down when shooting uncompressed RAW, to about 6.4 fps. The A7R IV also offers Mid and Lo Continuous modes rated at 6.0 and 3.0 fps respectively, however we did not test those modes."
I find the most reliable way to check FPS is to shoot your phone's stopwatch. Using the timestamps in EXIF may or may not be correct. I know people testing the R5's FPS ran into trouble relying on that method as the timestamps were only written when the file was written to card and not when it was taken. Once they tried the stopwatch method they were getting the advertised FPS. Maybe Sony isn't like that though and the timestamps referenced above are accurate??
If the EXIF data is correct, then it's 10x more accurate than using a stopwatch, as it contains millisecond information and it doesn't involve human error, and this error can easily be ~0.5 second which completely changes the resulting division.
Well, I just did a test:
1) I synchronized the A9's and my computer's clocks.
2) I just shot a burst from 15:48:30 => 15:48:40, 200 photos.
3) Buffer was cleared at 15:49:23 with 1 UHS-ii card.
(Note: those are the times I was observing on my computer clock)
Let's see what timestamps I get from the EXIF:
photo #1: 2021:02:21 15:48:30
photo #55: 2021:02:21 15:48:32
photo #200: 2021:02:21 15:48:40
Unfortunately, the A9 doesn't write the millisecond info, unlike the A1, so it means I got 200 shots in between 9.01.. and 10.99.. seconds, if we take the average of 10 seconds then it's 20fps. But you see that it could also be 10.5 seconds for example which would lead to 19fps. I just don't know because the milliseconds are missing.
As you can see, even the last shot was correctly timed, and the last image was written to the card at 15:49:23. So it means we can trust the EXIF information provided by Sony, and this is the most accurate measurement of the frame rate.
indusphoto wrote:
Has anyone else actually checked and verified that they can get 10fps on their camera? If so, what were the settings?
Measured my A7R4 in Hi+.
Compressed RAW, release priority, mechanical shutter, EFCS
Photographed digital stopwatch.
Short bursts to be sure not to fill buffer.
17 frame burst, 1.69 seconds. 10.06 fps.
silent shutter
10 frame burst, 1.38 sec, 7.25 fps
When I say 17 frame burst I mean the time from the first frame (frame 0) to the last frame (frame 17). There are actually 18 frames. Same for the 10 frame burst, there are actually 11 frames.
dclark wrote:
Measured my A7R4 in Hi+.
Compressed RAW, release priority, mechanical shutter, EFCS
Photographed digital stopwatch.
Short bursts to be sure not to fill buffer.
17 frame burst, 1.69 seconds. 10.06 fps.
silent shutter
10 frame burst, 1.38 sec, 7.25 fps
When I say 17 frame burst I mean the time from the first frame (frame 0) to the last frame (frame 17). There are actually 18 frames. Same for the 10 frame burst, there are actually 11 frames.
Dave
Thank you. I tried in full mechanical (I have left EFCS off) and Silent Shutter. I will check with EFCS shutter.
dclark wrote:
Measured my A7R4 in Hi+.
Compressed RAW, release priority, mechanical shutter, EFCS
Photographed digital stopwatch.
Short bursts to be sure not to fill buffer.
17 frame burst, 1.69 seconds. 10.06 fps.
silent shutter
10 frame burst, 1.38 sec, 7.25 fps
When I say 17 frame burst I mean the time from the first frame (frame 0) to the last frame (frame 17). There are actually 18 frames. Same for the 10 frame burst, there are actually 11 frames.
octo wrote:
If the EXIF data is correct, then it's 10x more accurate than using a stopwatch, as it contains millisecond information and it doesn't involve human error, and this error can easily be ~0.5 second which completely changes the resulting division.
Well, I just did a test:
1) I synchronized the A9's and my computer's clocks.
2) I just shot a burst from 15:48:30 => 15:48:40, 200 photos.
3) Buffer was cleared at 15:49:23 with 1 UHS-ii card.
(Note: those are the times I was observing on my computer clock)
Let's see what timestamps I get from the EXIF:
photo #1: 2021:02:21 15:48:30
photo #55: 2021:02:21 15:48:32
photo #200: 2021:02:21 15:48:40
Unfortunately, the A9 doesn't write the millisecond info, unlike the A1, so it means I got 200 shots in between 9.01.. and 10.99.. seconds, if we take the average of 10 seconds then it's 20fps. But you see that it could also be 10.5 seconds for example which would lead to 19fps. I just don't know because the milliseconds are missing.
As you can see, even the last shot was correctly timed, and the last image was written to the card at 15:49:23. So it means we can trust the EXIF information provided by Sony, and this is the most accurate measurement of the frame rate....Show more →
When I mentioned stopwatch I was using the one on my iPhone that has milliseconds.
indusphoto wrote:
Thank you. I tried in full mechanical (I have left EFCS off) and Silent Shutter. I will check with EFCS shutter.
The camera is definitely and unsurprisingly slower without EFCS.
