-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Comparison of GoPro Hero 7 Black and Sony X3000 Stabilization
- Thread starter Markr041
- Start date
sludge21017
Veteran Member
- Messages
- 4,624
- Solutions
- 8
- Reaction score
- 1,424
The GP is smoother, but those colors on the GP are a little excessive.
...for those unfamiliar.
As in ProTune / Flat Color plus white balance. While you're at it, turn sharpness down too, for those editing later on. I like to use some exposure compensation and cap ISO too.
My first impression of Hero 7 "gopro color" is that it is different than Hero 6 "gopro color".
I want to look into that further.
Note that you should use different ISO strategies for hypersmooth or not. And turn it off for low light captures or slowish panning
Various other things to keep in mind, still learning this thing (just got it).
As in ProTune / Flat Color plus white balance. While you're at it, turn sharpness down too, for those editing later on. I like to use some exposure compensation and cap ISO too.
My first impression of Hero 7 "gopro color" is that it is different than Hero 6 "gopro color".
I want to look into that further.
Note that you should use different ISO strategies for hypersmooth or not. And turn it off for low light captures or slowish panning
Various other things to keep in mind, still learning this thing (just got it).
Yes. This was shot in Protune with GoPro color, as I did not want to play with colors. I shot with sharpness low, but upped sharpness in post. Normally I shoot Protune flat and adjust color in post. I would also shoot in 4K 60p, but wanted to compare to the Sony, which doesnt have that option. Also, WB was set to 6500 for a cloudy day, and auto may have done better.
Anyway, the purpose is about stabilization. This is NOT an example of how I would use the GoPro.
Yes, but normally one would shhot jn Protune flat and then do color and contrast to taste. The settings were just for comparability, not to get the best out of the GoPro.
The video is to reveal the weakness of two-axis BOSS.
It's an overall interesting comparison. NTL I'm puzzled about the presence of a general sense of flutter to both samples. It's my understanding that walking adds a lot of challenges to stabilization. Maybe that's what's seen in the video.
I've posted this clip before, but take a look at this X3000 OIS&EIS/EIS/No IS (each segment runs about 5 minutes) comparison. It doesn't show the same flutter, even from a bike being banged by the road (watch the top of the windscreen relative to the landscape).
I've posted this clip before, but take a look at this X3000 OIS&EIS/EIS/No IS (each segment runs about 5 minutes) comparison. It doesn't show the same flutter, even from a bike being banged by the road (watch the top of the windscreen relative to the landscape).
sludge21017
Veteran Member
- Messages
- 4,624
- Solutions
- 8
- Reaction score
- 1,424
Being attached to ones helmet is a lot different than hand held or on a stick (not sure which you did).
sludge21017
Veteran Member
- Messages
- 4,624
- Solutions
- 8
- Reaction score
- 1,424
Durianrider would say you need a LUT. And that GoPro uses LUTs in 99% of their footage to get it to look good.
I'd post the video, but him and his model were out swimming, and pushing family safe for here.
I'd post the video, but him and his model were out swimming, and pushing family safe for here.
...in fact as much as you can, do trial clips using different mounts and strategies.
On GoPro's hypersmooth stabilization I wonder how the same clip would look at 4K60, or in some other modes.
I'm getting a feel for just what GoPro is doing here, for hypersmooth (which ties to the comment above).
For now, a clue is Abe's suggestion that for a Hero 5/6 with EIS he uses iso min 100 iso max 100 for outdoor shots.
But for the Hero 7, he revised this to iso min 100, iso max 400 for outdoor shots.
And no hypersmooth for low light captures, or where the pans are important.
On GoPro's hypersmooth stabilization I wonder how the same clip would look at 4K60, or in some other modes.
I'm getting a feel for just what GoPro is doing here, for hypersmooth (which ties to the comment above).
For now, a clue is Abe's suggestion that for a Hero 5/6 with EIS he uses iso min 100 iso max 100 for outdoor shots.
But for the Hero 7, he revised this to iso min 100, iso max 400 for outdoor shots.
And no hypersmooth for low light captures, or where the pans are important.
Last edited:
This thread is about gopro vs. Sony not about playing with settings on the gopro. That is interesting and important, deserving another thread.
Are you satisfied that we now know that BOSS is inferior to the latest gopro eis? That is the question here, not optimal ISO, frame rates, color, white balance, gamma for the gopro. Those are great topics, but not the topic here.
As in "to move in quick, irregular motions". Make my description the motion as "very small but noticeable flutter". As in, there's no gross, large motion, but neither is the presentation landscape totally still in the frame. It's present in both samples.
Also, for a 4K video, there are tons of processing artifacts (mostly large areas of blurring with little detail), notably in the darker bushes and trees along the path. I suspect this is due to YouTube processing, but I'm surprised by that. The clip I cited was shot at 1080, but transcoded to 4K, specifically minimize damage from YouTube processing. What's up on YouTube is quite close to the original raw footage.
The camera was helmet mounted. To get a sense of the overall camera motion, jump to about 12:00 for a sample of no image stabilization. Helmets are certainly stable on a rider (sad day when they're not!), but riders do get tossed around, taking the helmet along for the toss. In this case, the suspension was set for less than a full-on hard ride but firmer than average. Which means more motion from road irregularities.
Also, for a 4K video, there are tons of processing artifacts (mostly large areas of blurring with little detail), notably in the darker bushes and trees along the path. I suspect this is due to YouTube processing, but I'm surprised by that. The clip I cited was shot at 1080, but transcoded to 4K, specifically minimize damage from YouTube processing. What's up on YouTube is quite close to the original raw footage.
The camera was helmet mounted. To get a sense of the overall camera motion, jump to about 12:00 for a sample of no image stabilization. Helmets are certainly stable on a rider (sad day when they're not!), but riders do get tossed around, taking the helmet along for the toss. In this case, the suspension was set for less than a full-on hard ride but firmer than average. Which means more motion from road irregularities.
1. You dont *need* a LUT. A LUT is convenient, but if you know how to grade, you can do without it. And there are bad LUTs and good LUTs. They are just color grades by someone else.
2. You can use ACES. In Vegas Pro there is an input transformation from Protune flat native color to REC709 in ACES. That in principle is better than a LUT since it is an exact math transform not someone's else's color grade captured in a LUT.
But this has nothing to do with the topic of this thread, which is about stabilization on two cameras and their resolution. Of course, color is very important.
Last edited:
I used ISO min 100 and max 400, as was recommended. The ISO tie in is due to stabilization performing better at higher shutter speeds, so in lower light higher ISO is needed. In my video the light was pretty low, no sun, and I suspect the ISO was at the higher end, making the video softer due to aggressive noise reduction.
creaDVty
Senior Member
[No message]
sludge21017
Veteran Member
- Messages
- 4,624
- Solutions
- 8
- Reaction score
- 1,424
This is the very first FDR-X3000 hand held video I've seen. There's not a lot of walking - but some.
1 segment has a little too much shake though.
He might have stabilized some in post.
but no gimball.
sludge21017
Veteran Member
- Messages
- 4,624
- Solutions
- 8
- Reaction score
- 1,424
Speaking of quality - did you encode this at a low bit rate?
I know it wasn't a sunny day - but it should look much better than this.
I know it wasn't a sunny day - but it should look much better than this.
I am disappointed by resolution, but its no better on the Sony. Sony shot at the usual 100 Mbps, H264. The Gopro at 60 Mbps HEVC (about equivalent 120 Mbps H264). Rendered at 100 Mbps XAVC S, which is what I use for most of my videos. I
Similar threads
Keyboard shortcuts
- f
- Forum