A Brief History of Quality: Comparing BRAW 3:1, Q0, Q5

But I’ve seen the published data rates for Q0 and Q5 are not some rule in the algorithm but only a guideline for us to compare. Now we’re shooting Q0 and Q5, the ranges are actually greater than was published. And that’s good news.

Q3 is a good addition that would be used for sure, but believe me Q0 consumes less resources than I expected. There can be times where media storage is a concern and Q3 would likely be a welcome alternative for many.


Sent from my iPhone using Tapatalk

So Q0 on the Pocket 4K might not be good for panning shots, as it would introduce artifacts and such-like?

Interesting looking back on this thread given we now have Q1 and Q3 on the UMP12K.

Good question re panning. A slow pan should be good; a fast pan is usually over with very quickly and there’s likely no attempt to define edges as they are blurred.

Between Q0 and Q5, which handles movement within the shot better? Another user complained of mushy shots when using Q5. Is Q5 better for shots with static subject matters?

Todd, others may have a different view, but I think Q0 will give you better results than Q5 when there is motion in the frame or camera movement. I’m not the best person to ask as I’ve never done a comparison if Q0 to Q5 with motion.

Q5 can be a good choice if you need to economize on media or data rates. But I regularly use Q0 or 12bit ProRes 444. At one time I used to use 444 HQ, but that comes with a hefty premium. If I had the option, I’d use Q1 for the same reason.

The recording capacity of CFast 2 and SSD cards are quite large and the makes shooting Q0 or ProRes 444 feasible for many. I see some SSD media will be going to 4TB but I’d be nervous about using it unless certified by BMD.

The greater storage requirements and high data rate of Q0 on your RAID or other storage system is something to weigh as well. A documentary shooting something like 100:1 might dictate the Q0 high data rate media should be transcoded to a more compressed codec on the working drives (or you shoot in Q5). For me one of the beauties of shooting BRAW is the efficiency of using the media as is in DaVinci Resolve without transcoding to an optimized codec. But if I was doing the documentary I had planned to do (before my subject passed away), much of it would have been shot in Q5 just so I could store it on the edit computer.

No, Rick. It is not possible that motion has any influence on the codec as it compresses each frame independently - it is all-I, not GOP.

So it comes down how much detail you have in your framing - like a wide shot of a forest or meadow with lots of details. For shots with a lot of bokeh or motion blur Q5 should be good to go.

Robert, my thinking was the motion or camera movement that naturally will blur the details in each frame (due to relatively slow shutter speeds like 1/50th second) compared to a lockdown shot without motion that will show detail much better.

No, Rick. It is not possible that motion has any influence on the codec as it compresses each frame independently - it is all-I, not GOP.

So it comes down how much detail you have in your framing - like a wide shot of a forest or meadow with lots of details. For shots with a lot of bokeh or motion blur Q5 should be good to go.

I heard one user shot with Q5 and got a lot of mushy details with some motion in the shot. So, is Q0 or Q5 good for shots with camera movement or in shot motion?

I have shot a lot with Q5 and have no complaints about.

I have shot a lot with Q5 and have no complaints about.

Thanks. Do you think Q5 will be good for green screen shots?

For green screen I'd suggest using the highest quality your media can handle. You'll want every detail preserved as much as possible.
Some folks even shoot shorter exposure times for green screen and introduce artificial motion blur after compositing.

I agree with Uli — for green screen use the highest recording your media can handle. I'd also light the shot for ISO 400 on the G2 or the 12K (and ISO 200 on the 4K/6K Pocket) if possible to keep the noise low.

For the BMPCC4K, what setting would you recommend for shooting green screen when recording to the 500GB Samsung T5? (I guess I should have specified the camera.)

For the BMPCC4K, what setting would you recommend for shooting green screen when recording to the 500GB Samsung T5? (I guess I should have specified the camera.)

I'd shoot BRAW 3:1 or BRAW Q0 at the lowest ISO to which you are able to light the scene.

For the BMPCC4K, what setting would you recommend for shooting green screen when recording to the 500GB Samsung T5? (I guess I should have specified the camera.)

I shoot BRAW 3:1 or BRAW Q0 at the lowest ISO to which you are able to light the scene.

I have shot green screen at 3:1. I'll try Q0 next.

So it seems like I am resurrecting this thread. I want to speed up my playback for editing. If I record in q0, I have bigger files so they are harder to pass around, which would lead me to using q5 to save on disk space. But, my question is, is it easier to work in the timeline with q0 files since they are closer to uncompressed than q5? That is, would q0 decoding be easier than q5 therefore playback of q0 will be faster?

While BRAW is in general very efficient, Q0 playback requires more CPU + GPU power than Q5.

BMD has an app that will show you how many frames per second of each compression level + frame size that your computer can process — Blackmagic Raw Speed Test

It’s part of the package that installs with the BRAW SDK

https://www.blackmagicdesign.com/support/download/069ca1d5b24546d7b099e39365152121/Mac%20OS%20X

Since the Q compressions are variable rather than fixed compression ratios, the app doesn’t show Q levels. It only shows fixed compression ratios. But, if you run it you’ll see that the higher compression ratios result in more frames able to be processed per second, and the reverse for lower compression ratios.

[CaptainHook]

As Jamie says, in general the more compressed the Blackmagic RAW file is the less data there is to process. The less data to process, the less time it takes to process (i.e faster).

Blackmagic Raw Speed Test

It’s part of the package that installs with the BRAW SDK

the more compressed the Blackmagic RAW file is the less data there is to process

Thanks! Answered my query.