I really appreciate how thorough you were with these tests! And yes, all Canon's transform luts are trash. I don't understand what's wrong with their quality control on that front
Thanks Tyler! Love your stuff man. Yea the Canon LUT thing is so weird. It's as if they're releasing luxury cars with all these amazing features, but then they don't tell the owners what type of gas or oil to use to make the cars perform their best.
I have no problem matching R5, c300 mkII and C70 with Canon's LUTs as long as I custom WB all three. I have had problems with Canon LUTs when using raw material.
FYI One more thing. I just downloaded Canon’s newest LUTs (says v2.0 on each). They are not that new but I’ve been using v1.1 for way too long. For me these 2.0 LUTs seem to fix many problems. Interview shots from c70, c300mkii, Eos r, r5 and r6 match quite well as long as I custom WB all cameras manually. The LUTs also seem to fix the blue LED color shifting to violet which is part of the Canon color science. I am using 65x65 3D LUTs with the full to full range. Interestingly Canon omitted the full to video LUTs from the new update. All this essentially means that for future projects I am more likely to use Canon LUTs because currently I’ve been mostly using CSTs in Resolve as I like starting point better. CSTs just seems a bit more taxing on the computer. (for those who don’t know) CST= Color Space Transform
The lines of the vectroscope “opening up” with the .71 adapter means that it’s introducing more colors around it. The colors must be fringing a bit due to the adapter.
Regarding All-I vs Long GOP, testing a shot where the tripod is locked and the subject/background is nearly identical, is not helpful to see the benefits of All-I. Remember, since little in the image is changed except for her slight lifting/turning of the model's head, then long GOP is great for this particular scenario. If, however, it were a busy shot like a camera moving on a steadicam through a detailed background environment like brick walls, a busy street, or even along the water, then one can see long GOP start to fall apart in compression compared to All-I. If all one is doing is filming an interview, for example, then long GOP is perfectly fine. Lots of movement & action, however, All-I would help.
This was the most thorough review of the C 300 and C 70 and you were even able to find out there the reasons for the Problems are. Another reviewer suggested to not use the Canon RF to EF adapter, but since i don’t have both Cameras i can’t prove or test his claim......definitely outstanding work
C70 owner here. Great video Chris - have sub'd. I have customised a button to switch quickly between 25 and 120 fps in certain circumstances - but could only find a way to do this if both are long-GOP. Although initially concerned about giving-up quality at 25 fps - real-world - I cannot see any difference between the codecs and am beyond impressed with the look and DR of this camera
As an optician I think that's the reason: The Speedbooster lens has a coated surface to reduce reflection. It's impossible to reduce all refection, so the coating has the thickness to reduce mostly blue. Because of that, more red will enter the sensor. So we have to live with it, still better than no coating and losing light ;). Thanks for the test.
Thanks, Chris! I found it helpful to see where the c70 and c300 mkiii differ. Also, it is good to see you back! Would love to see more project breakdowns as well as industry tips/tricks. Keep up the awesome content!
Thank you for taking the time. This was very in-depth and very much what I do to test new cameras. I did have the c70 for few days but did not manage to do a chart based test for matching (only real world scenario with me standing in front). I think the C70 is an easy match to the C300MKII and EOS R5,R6 and EOS R as long as one uses custom WB on all cameras and choses the Neutral color matrix. Also good point that same sensor does not mean same processing... I think many get to caught up in the marketing lingo :D. As far as I know the beauty of Alexa, Mini and Amira is that they not only use the same ALEV III sensor but the processing is exactly the same ;).
@@ChrisFrancis reminded me of a Gerald Undone video. I've been filming for 15yrs but listening to guys like you reminds me there's still plenty to learn.
I think All-I vs LongGOP is a bit misunderstood where people are like All-I - good, LongGOP - bad. The data rate allocation is really important. All-I needs a much higher data ceiling to encode each frame with integrity. I don't know the actual figures (and it depends on how each format is implemented in any case) but you should be looking for data rates of around 4 times to 8 times higher for an All-I format to be outperforming an equivalent LongGOP one. On most camera systems the data rate afforded ALL-I, being only around twice the amount allocated to the LongGOP or so, is not enough to allow it to exceed the more efficient format. Generally speaking you should not expect to see an image quality increase using ALL-I (unless the data rate difference is very high). ALL-I can give big advantages with moving detail though, despite the results of your test here. It really depends how much stuff is actually moving in frame / how fast / etc. It's worth using when the content demands it.
That is why it is recommended that if you need an exact camera match to order two of the same cameras directly one after the other at the same manufacturing facility bc these issues are not exclusive to different camera bodies.
new to the cinema camera universe, but this was definitely interesting to learn about regarding issues with shooting and matching cameras, so thanks for making video.
Not hard to match them in post if you shoot a color chart and/or white balance chart, but for sure a little annoying that they don't match exactly. If I end up buying the C70 as a B-CAM I'll definitely run it through the same tests to confirm the color biases and then will probably tweak the in-camera color settings so I can get them to match right in the camera.
Wow, there was a ton of noise in all of them for just being at the base ISO and lit. So much for DGO lol. The RAW looked much richer. Also really hate magenta, which the C70 has a ton of, I also noticed it didn't have as good of highlight rolloff when shooting outside next to the C300MKiii. I think Canon held back on the C70.
Great comparison! Just one fatal flaw at the end: on a mostly white backdrop, the 160mbps is going to be very similar to the 410mbps. You need to do a comparison that shows something like falling rain, waves, foliage, etc to see the difference and I bet it's pretty drastic!
For sure. I have ended up shooting in the higher data rate 99% of the time as I no longer handle post production on my shoots so storage space is now longer a consideration If I was still creating a lot of content for RUclips I would certainly use the long GOP to save on storage
Hi Chris, usually on RUclips I can't see the differences with many testers results, but here I can. It might be even clearer on Vimeo which I understand has less aggressive compression. I'm surprised at the difference between apparently identical non cinematic lenses and the effect of the introduction of the 0.71 x glass was not something I'd have thought of either, so all in all a very useful and well presented lesson in colour balancing for an amateur film maker and thank you very much for all the trouble. A shout out to your talent Khrista as well and I do hope you are all safe and well. The dance interlude gave me a nice smile. Liked and subd. BobUK
Great video!! I own both as well and was curious about this because I sometimes prefer the c70 to get the wider FOV with the adapter to not crop my lenses. One thing I would say about switching to lower bit rate though is be careful because this is a controlled light ideally lit kind of setting. The point of the data in my view is for when you have to push the footage harder in your grading for less ideal situations and broader exposure ranges. I think in any more challenging situations like that there might be more of a difference. I just wouldn't make a decision across the board based on this circumstance to change the bitrate like that. Thanks again!!
Thank you for these great samples. Subscribed. I hear you about Long Hop Vs All-I - the only quality difference I've found in my work on XT'4s is the lower quality of life while waiting for All-I to download, backup or render. Cheers.
Amazing test, very thorough and I appreciate you took great and proper care at ensuring that the test was not biased towards any of the cameras. I’m in the market for my first cinema camera and I’m already 99% certain I’m getting this camera, so this was helpful!
Very thorough and helpful as I have both of these cameras. In practical terms for the work I do... the cameras match just fine. I might preset the C70 with a setting that brings the camera closer to the C300m3 but that's getting pretty deep into the camera nitty gritty. Like you say, different lenses probably have more effect.
I love all the tests but I am also really interested in the lighting setups as well. Any chance you could do a video deconstructing the setup with the girl reading a book? The decision to use an 8x8 bounce vs a large umbrella for example. Thanks!
Really appreciate this, Chris. Wonderful test. Is there more information regarding the color matrix options. I didn't even know about those settings. How do they impact the image?
@@ChrisFrancis I want to get a camera to make indie film. So I'm not sure which one is better considering I want to save money and both seem to shoot almost the same quality.
Very nice and in depth! Seems that the c70 could be matched and balanced in camera using the color / phase adjustments. I'd be curious if the c70 matched the other c70 without the speedbooster. If you're thinking of long GOP recording, the HEVC long gop has way less macro blocking than the XF-AVC long gop because it's h265 encoding. Just harder to edit.
Yea, if/when I buy a C70 as a B-Cam I'll run it through these test and then tweak the internal color settings to try to get it to match as closely as possible in-camera (which I agree should be possible to get it dialed in close enough).
Informative! Great comparison really showing what is going on with the c70 and c300mk3. This video is simply ja cleverly put together. Amazing! Canon needs to get those transform luts updated asap.
The amount of time and effort in this test is crazy. Thanks so much! I can't afford any of these high end equipment just yet, but it's interesting to learn more. One question though, do you find the internal NDs of Canon to be kinda limiting? Correct me if I'm wrong but I think they're kind of like in fixed steps and not a internal VND like the FX6 or FX9 right? Would that be an issue that forces you to use non-native ISO in order to compensate for 1/3 stop differences that the fixed internal NDs might not cover? Oh, and you might have gotten Mbps and MBps confused.
You’re welcome! No I don’t find the ND steps limiting, that’s basically how it’s always been done. Great quality variable ND filters didn’t exist until a few years ago actually so they’re still a relatively new thing. Personally I don’t have a problem adjusting the iso up or down a little bit to nail exposure. Especially when bumping it up, the Canon image is so clean when shooting compressed that it doesn’t really add any noticeable noise of just bumping a little
An over abundance of caution to eliminate any variances in the tests. In general black balances will be most accurate when the camera is at its operating temperature. When dealing with cameras like RED this more important because the camera gets so hot when recording.
You will only see a difference between All-I and Long Gop in a more stressful situation for the codec. A static shot with a plain background is the best situation for long Gop 😊 It might be different if you follow a Mountainbike through the woods.
For sure, that’s why I had her shake her hair so much towards the end of the test. I’ve been told that motion in fine detail like hair or grass is where you can start to notice things, but I’m not seeing anything in the moving hair. Will for sure do more tests on All-I vs. Long GOP on these cameras.
I was literally JUST going through the same annoyances with the CLOG2 to rec709 LUT. It does make it darker for sure, and kinda clips the blacks. Maybe this is one of those scenarios where we need to overexposed by a stop when shooting CLOG2?
Amazing video! Thanks for this! Curious why you chose the 65 version of the luts instead of 33? I was under the impression 33 was for 10 bit footage while 65 was for 12 bit? Also thanks to your patient model! My fiancé hates posing for my own camera tests! 😃
Thanks Eric. I chose the 65 versions because my friend that's a professional colorist told me to always use the 65 versions when available, simple as that, ha. Also, I'll pass along your thanks to my VERY patient wife 😂
I agree. The C70 has a slightly more processed look, extra contrast a part of that. Years ago, I was comparing the Nikon D90 and the Nikon D300. Both had the same (crop) sensor. However, the D300 had a better image as it’s processing was better, so I bought the D300. In a similar way, I feel the internal processing is better in the C300 Mark 3 than the C70. I preferred the former’s image, but it is more expensive. Though close, having the same sensor, doesn’t mean it has the same output, in my view.
Man if you think two totally different cameras being that close to identical in log is not matching, then matching doesn’t exist. Go on a tv truck show and match a ikegami with a sd lens and a sony with a hd lens, then you can come back and say whats “matching” 😂. The difference isn’t them not matching its the tolerance difference between sensors on the manufacturing line. They are the same. At least as same as the micro millimeter machines can make em. To be fair its good you saw the 1 percent differences, but this has nothing to do with the cameras, you set up 4 sony f5’s and they’ll be sometimes even more off. No two cameras in log will match closer. Its not possible. Good lesson to learn.
Curious, why didn’t you just white balance the cameras to your color card? This would have solved the majority of the discernible differences between the cameras - even the lens variable.
I really admire the work that you put in here. However it would’ve been very helpful if you could also test how the speed booster affects the image quality in terms of details and chromatic aberration, etc. Thanks
Really good video thanks for putting this together. So essentially, your paying nearly double for an EVF, bigger touchscreen and the form factor. Colour is negligible on the grand scheme of things.
I’d also add RAW and full size XLRs to that list, but yea, picking the C300MKIII over the C70 is mostly for the more robust and versatile cinema camera body (which personally I would pick the C300MKIII over the C70 9 times out of 10). It really depends on what type of work you do and your personal preferences, but if you’re looking for the most bang for your buck strictly on image quality the C70 value is about as good as it gets
The sensor is only half of it. How the camera uses that sensor matters. Many cars share the same engine, but how you tune that engine can produce double the power
I really appreciate how thorough you were with these tests! And yes, all Canon's transform luts are trash. I don't understand what's wrong with their quality control on that front
Thanks Tyler! Love your stuff man. Yea the Canon LUT thing is so weird. It's as if they're releasing luxury cars with all these amazing features, but then they don't tell the owners what type of gas or oil to use to make the cars perform their best.
@@ChrisFrancis I like the making a lemonade documentary
@@isaacasunciongallardo9781 Thanks Issac!!
I have no problem matching R5, c300 mkII and C70 with Canon's LUTs as long as I custom WB all three. I have had problems with Canon LUTs when using raw material.
FYI
One more thing. I just downloaded Canon’s newest LUTs (says v2.0 on each). They are not that new but I’ve been using v1.1 for way too long.
For me these 2.0 LUTs seem to fix many problems. Interview shots from c70, c300mkii, Eos r, r5 and r6 match quite well as long as I custom WB all cameras manually.
The LUTs also seem to fix the blue LED color shifting to violet which is part of the Canon color science.
I am using 65x65 3D LUTs with the full to full range. Interestingly Canon omitted the full to video LUTs from the new update.
All this essentially means that for future projects I am more likely to use Canon LUTs because currently I’ve been mostly using CSTs in Resolve as I like starting point better. CSTs just seems a bit more taxing on the computer.
(for those who don’t know)
CST= Color Space Transform
Thanks for the shout-out and was happy to have my cameras be put to the test! Learned a bunch from this video.
The lines of the vectroscope “opening up” with the .71 adapter means that it’s introducing more colors around it. The colors must be fringing a bit due to the adapter.
Regarding All-I vs Long GOP, testing a shot where the tripod is locked and the subject/background is nearly identical, is not helpful to see the benefits of All-I. Remember, since little in the image is changed except for her slight lifting/turning of the model's head, then long GOP is great for this particular scenario. If, however, it were a busy shot like a camera moving on a steadicam through a detailed background environment like brick walls, a busy street, or even along the water, then one can see long GOP start to fall apart in compression compared to All-I. If all one is doing is filming an interview, for example, then long GOP is perfectly fine. Lots of movement & action, however, All-I would help.
Best part is the guy dancing at the end. More of him in camera tests pls.
The lines were fanning out because of chromatic abberation on the edges of the colour patches 😉
This was the most thorough review of the C 300 and C 70 and you were even able to find out there the reasons for the Problems are. Another reviewer suggested to not use the Canon RF to EF adapter, but since i don’t have both Cameras i can’t prove or test his claim......definitely outstanding work
thank you!!
C70 owner here. Great video Chris - have sub'd. I have customised a button to switch quickly between 25 and 120 fps in certain circumstances - but could only find a way to do this if both are long-GOP. Although initially concerned about giving-up quality at 25 fps - real-world - I cannot see any difference between the codecs and am beyond impressed with the look and DR of this camera
for sure, the long GOP really exceeded my expectations. thanks for the sub!
This is really a very in-depth comparison i have ever seen for C70, C70+.71 booster, and C300MKIII... good work you have done!! Great thanks!!
As an optician I think that's the reason: The Speedbooster lens has a coated surface to reduce reflection. It's impossible to reduce all refection, so the coating has the thickness to reduce mostly blue. Because of that, more red will enter the sensor. So we have to live with it, still better than no coating and losing light ;). Thanks for the test.
awesome insight, thanks for sharing!
Thanks for taking the time to do this so thoroughly
A lot of work here! Thanks so much for sharing all this useful information! Proud to be your subscriber!
This confirmed many of the things I know but I’m very glad you made it
The DIT community thanks you for your service.
Thanks, Chris! I found it helpful to see where the c70 and c300 mkiii differ. Also, it is good to see you back! Would love to see more project breakdowns as well as industry tips/tricks. Keep up the awesome content!
Thanks @Isaiah! Really appreciate the encouragement
Thank you for taking the time. This was very in-depth and very much what I do to test new cameras. I did have the c70 for few days but did not manage to do a chart based test for matching (only real world scenario with me standing in front). I think the C70 is an easy match to the C300MKII and EOS R5,R6 and EOS R as long as one uses custom WB on all cameras and choses the Neutral color matrix.
Also good point that same sensor does not mean same processing... I think many get to caught up in the marketing lingo :D. As far as I know the beauty of Alexa, Mini and Amira is that they not only use the same ALEV III sensor but the processing is exactly the same ;).
Great job comparing and discovery relative to the importance of lenses. Thanks Chris
Absolute gold. I really appreciate these test. 🙌🏼
That was a deep dive! Thanks for the comparison.
Phew! I need a vacay after that one 😅. Glad you liked it!
@@ChrisFrancis reminded me of a Gerald Undone video. I've been filming for 15yrs but listening to guys like you reminds me there's still plenty to learn.
@@Extremehick Awesome, thanks James!! I'm always learning, too! (who knew photo lenses could vary so much in quality?!)
I think All-I vs LongGOP is a bit misunderstood where people are like All-I - good, LongGOP - bad. The data rate allocation is really important. All-I needs a much higher data ceiling to encode each frame with integrity. I don't know the actual figures (and it depends on how each format is implemented in any case) but you should be looking for data rates of around 4 times to 8 times higher for an All-I format to be outperforming an equivalent LongGOP one.
On most camera systems the data rate afforded ALL-I, being only around twice the amount allocated to the LongGOP or so, is not enough to allow it to exceed the more efficient format.
Generally speaking you should not expect to see an image quality increase using ALL-I (unless the data rate difference is very high).
ALL-I can give big advantages with moving detail though, despite the results of your test here. It really depends how much stuff is actually moving in frame / how fast / etc. It's worth using when the content demands it.
Such an awesome review. Much appreciated
awesome video - insanely informtive not sure why more people havn't shown this comparison online with videos already - very eye opening
Whew. LOTS of work, LOTS of time. Thanks for your due diligence!
Thanks mate, extremely useful now we have both
Incredibly thorough tests. Amazing!
That is why it is recommended that if you need an exact camera match to order two of the same cameras directly one after the other at the same manufacturing facility bc these issues are not exclusive to different camera bodies.
How will you accomplish that?
Excellent, exhaustive comparisons - some great takeaways and tips for ALL videographers regardless of what camera you shoot.
This is a great comparison! Thank you man!
Hey Chris,
thanks for this Comparison. It was very informative.
new to the cinema camera universe, but this was definitely interesting to learn about regarding issues with shooting and matching cameras, so thanks for making video.
Great video Chris. Super detailed and well done. Thanks for sharing!
Thanks Griffin! Really enjoying your vids
Great video. I picked up a C70 with the option of renting a C300iii for different projects as the need arises. Good to know it's not actually 1:1.
Not hard to match them in post if you shoot a color chart and/or white balance chart, but for sure a little annoying that they don't match exactly. If I end up buying the C70 as a B-CAM I'll definitely run it through the same tests to confirm the color biases and then will probably tweak the in-camera color settings so I can get them to match right in the camera.
Great comparison.. Thumbs up too you.
Thanks Chuck!
Wow, there was a ton of noise in all of them for just being at the base ISO and lit. So much for DGO lol. The RAW looked much richer. Also really hate magenta, which the C70 has a ton of, I also noticed it didn't have as good of highlight rolloff when shooting outside next to the C300MKiii. I think Canon held back on the C70.
Great points
Great comparison! Just one fatal flaw at the end: on a mostly white backdrop, the 160mbps is going to be very similar to the 410mbps. You need to do a comparison that shows something like falling rain, waves, foliage, etc to see the difference and I bet it's pretty drastic!
Actually from the manual I think the C70 is 260mbps at 60fps and the C300mkIII is 810mbps, so it's a huge data rate difference
For sure. I have ended up shooting in the higher data rate 99% of the time as I no longer handle post production on my shoots so storage space is now longer a consideration
If I was still creating a lot of content for RUclips I would certainly use the long GOP to save on storage
Hi Chris, usually on RUclips I can't see the differences with many testers results, but here I can. It might be even clearer on Vimeo which I understand has less aggressive compression. I'm surprised at the difference between apparently identical non cinematic lenses and the effect of the introduction of the 0.71 x glass was not something I'd have thought of either, so all in all a very useful and well presented lesson in colour balancing for an amateur film maker and thank you very much for all the trouble. A shout out to your talent Khrista as well and I do hope you are all safe and well. The dance interlude gave me a nice smile. Liked and subd. BobUK
Thanks Bob, glad to hear it was helpful. Appreciate you taking the time to comment.
Gracias por ofrecer más claridad sobre estas dos cámaras , Justo estoy en el proceso de comprar una de las dos
The camera's data rates you're stating are mega "bits" per second (Mbps), NOT mega "bytes" (MBps.)
Helpful video btw. Thanks!
you are correct, i always get bytes and bits confused
damn. even when watching a camera comparison I can still learn a ton
A lot of work here! Thanks so much for sharing all this useful information! Can’t wait for part 2! Proud to be your subscriber!
Decide for me. I can’t decide if I should undergo the raw workflow on the c300 or just use the C 70. Help !
You don't need to record for 30 minutes in order to black balance. Leaving the camera on for 20 min is enough to do the black balance. Good video.
Great research, helpful results!!! Subscribed ✔️
Thanks Mark!
5:15 Chromatic aberration is quite obvious with the 0.71x adapter.
Loved this comparison! Been debating wether to get the C70 or C300mkiii and I feel Im good with C70! Thanks!
Fantastic test, lots of useful info for a DP
Great video!! I own both as well and was curious about this because I sometimes prefer the c70 to get the wider FOV with the adapter to not crop my lenses. One thing I would say about switching to lower bit rate though is be careful because this is a controlled light ideally lit kind of setting. The point of the data in my view is for when you have to push the footage harder in your grading for less ideal situations and broader exposure ranges. I think in any more challenging situations like that there might be more of a difference. I just wouldn't make a decision across the board based on this circumstance to change the bitrate like that. Thanks again!!
Thanks! Would like to see an update including the C70's raw :)
Hey Chris, congrats on the awesome video
Thank you for creating such a great video
Very detailed comparison. I wonder how the standard canon EF-RF adapter would compare to the changes seen in the 0.71 adapter.
@joaquin, the standard EF-RF adapter is what was used on the other C70 featured in these tests
@@ChrisFrancis Ah, my mistake! Thanks for taking the time to do this.
@@joaquinbarlow No worries!, Thx!
I forget if I commented on this video, but I love this video!
Excellent comparison. Nice work!
Thank you for these great samples. Subscribed. I hear you about Long Hop Vs All-I - the only quality difference I've found in my work on XT'4s is the lower quality of life while waiting for All-I to download, backup or render. Cheers.
Nice Video! But the chromatic aberration with the speedbooster does not look good. Do you think it's coming from the lens or the adapter?
Great job on this Chris! Some solid info. Thank you.
Excellent test!
Wow great test ! Thanks
Just a note that the data rates are actually megabits, not megabytes - 8 megabits to 1 megabyte, thanks for doing the tests!
Amazing test, very thorough and I appreciate you took great and proper care at ensuring that the test was not biased towards any of the cameras. I’m in the market for my first cinema camera and I’m already 99% certain I’m getting this camera, so this was helpful!
Thank you, glad to hear it was helpful for you. Good luck with your upcoming purchase!
Very thorough and helpful as I have both of these cameras. In practical terms for the work I do... the cameras match just fine. I might preset the C70 with a setting that brings the camera closer to the C300m3 but that's getting pretty deep into the camera nitty gritty. Like you say, different lenses probably have more effect.
I love all the tests but I am also really interested in the lighting setups as well. Any chance you could do a video deconstructing the setup with the girl reading a book? The decision to use an 8x8 bounce vs a large umbrella for example. Thanks!
It's 410 megabits, not megabytes. Difference is only 8 times
Really appreciate this, Chris. Wonderful test. Is there more information regarding the color matrix options. I didn't even know about those settings. How do they impact the image?
Funny you should ask, I literally just hit export on my next video “Canon Color Matrixes Explained”. Should go live on Monday 👍🏻
Your video is very helpful and just makes me wonder why spend the money on the C300 if C70 is just as good?
All depends on the environment you use it in. Both are great cameras
@@ChrisFrancis I want to get a camera to make indie film. So I'm not sure which one is better considering I want to save money and both seem to shoot almost the same quality.
@@atomicdynamo6062 C70 then spend the rest of the money getting the best actors possible and a good sound person
Brilliant comparison + excellent video!!
Would love to see some Canon R5 4K HQ thrown into some comparisons with C70 etc! :)
Very nice and in depth! Seems that the c70 could be matched and balanced in camera using the color / phase adjustments.
I'd be curious if the c70 matched the other c70 without the speedbooster.
If you're thinking of long GOP recording, the HEVC long gop has way less macro blocking than the XF-AVC long gop because it's h265 encoding. Just harder to edit.
Yea, if/when I buy a C70 as a B-Cam I'll run it through these test and then tweak the internal color settings to try to get it to match as closely as possible in-camera (which I agree should be possible to get it dialed in close enough).
Informative! Great comparison really showing what is going on with the c70 and c300mk3. This video is simply ja cleverly put together. Amazing! Canon needs to get those transform luts updated asap.
Really thorough, I'll pass this video on to my guys.
The amount of time and effort in this test is crazy. Thanks so much! I can't afford any of these high end equipment just yet, but it's interesting to learn more.
One question though, do you find the internal NDs of Canon to be kinda limiting? Correct me if I'm wrong but I think they're kind of like in fixed steps and not a internal VND like the FX6 or FX9 right?
Would that be an issue that forces you to use non-native ISO in order to compensate for 1/3 stop differences that the fixed internal NDs might not cover?
Oh, and you might have gotten Mbps and MBps confused.
You’re welcome! No I don’t find the ND steps limiting, that’s basically how it’s always been done. Great quality variable ND filters didn’t exist until a few years ago actually so they’re still a relatively new thing. Personally I don’t have a problem adjusting the iso up or down a little bit to nail exposure. Especially when bumping it up, the Canon image is so clean when shooting compressed that it doesn’t really add any noticeable noise of just bumping a little
Your video is great.I want to be consulted.How does backlight behave 0.71, Is glare serious?
Chris: I thought same censor and same color science
Canon: you just got played son. :D
Why did you wait for 30 minutes before running the auto black balance?
An over abundance of caution to eliminate any variances in the tests. In general black balances will be most accurate when the camera is at its operating temperature. When dealing with cameras like RED this more important because the camera gets so hot when recording.
That's an outstanding job! Thank you so much!
You will only see a difference between All-I and Long Gop in a more stressful situation for the codec. A static shot with a plain background is the best situation for long Gop 😊 It might be different if you follow a Mountainbike through the woods.
For sure, that’s why I had her shake her hair so much towards the end of the test. I’ve been told that motion in fine detail like hair or grass is where you can start to notice things, but I’m not seeing anything in the moving hair. Will for sure do more tests on All-I vs. Long GOP on these cameras.
Now that's a real test 👍
Brilliant, thank you.
Thanks Chris! Glad it was helpful.
Great content ! Very informative
TIGHT REVIEW!
Would you trust the c70 as your main camera for a documentary ? Thanks for the review
Yep, but I'd for sure want to figure out an ergonomic Zacuto Z-finder configuration for it for sure.
Great information. Thanks.
Very good comparisons
Thanks!
I was literally JUST going through the same annoyances with the CLOG2 to rec709 LUT. It does make it darker for sure, and kinda clips the blacks. Maybe this is one of those scenarios where we need to overexposed by a stop when shooting CLOG2?
Amazing video! Thanks for this! Curious why you chose the 65 version of the luts instead of 33? I was under the impression 33 was for 10 bit footage while 65 was for 12 bit? Also thanks to your patient model! My fiancé hates posing for my own camera tests! 😃
Thanks Eric. I chose the 65 versions because my friend that's a professional colorist told me to always use the 65 versions when available, simple as that, ha. Also, I'll pass along your thanks to my VERY patient wife 😂
The C70 just looks more contrasty. The C300 Mark 3 seems to have smoother color transitions in the tones.
I agree. The C70 has a slightly more processed look, extra contrast a part of that. Years ago, I was comparing the Nikon D90 and the Nikon D300. Both had the same (crop) sensor. However, the D300 had a better image as it’s processing was better, so I bought the D300. In a similar way, I feel the internal processing is better in the C300 Mark 3 than the C70. I preferred the former’s image, but it is more expensive. Though close, having the same sensor, doesn’t mean it has the same output, in my view.
I'm new to filming and would like to be your student. I like this video a lot.
Man if you think two totally different cameras being that close to identical in log is not matching, then matching doesn’t exist.
Go on a tv truck show and match a ikegami with a sd lens and a sony with a hd lens, then you can come back and say whats “matching” 😂.
The difference isn’t them not matching its the tolerance difference between sensors on the manufacturing line. They are the same. At least as same as the micro millimeter machines can make em.
To be fair its good you saw the 1 percent differences, but this has nothing to do with the cameras, you set up 4 sony f5’s and they’ll be sometimes even more off. No two cameras in log will match closer. Its not possible. Good lesson to learn.
let's go safemoon, doge. i want me a c70. LOL
😂
This was helpful bud. I'm a C70 fanboy.
Curious, why didn’t you just white balance the cameras to your color card? This would have solved the majority of the discernible differences between the cameras - even the lens variable.
Is it just me or does the rendering of the face look slightly different between the c300 and c70. like blockier on the c70
Great informative video.
Did Canon in fact update its look-up-tables?
I really admire the work that you put in here. However it would’ve been very helpful if you could also test how the speed booster affects the image quality in terms of details and chromatic aberration, etc. Thanks
That's a great idea for a follow up video, thanks!
Love this thank you.
Really good video thanks for putting this together. So essentially, your paying nearly double for an EVF, bigger touchscreen and the form factor. Colour is negligible on the grand scheme of things.
I’d also add RAW and full size XLRs to that list, but yea, picking the C300MKIII over the C70 is mostly for the more robust and versatile cinema camera body (which personally I would pick the C300MKIII over the C70 9 times out of 10). It really depends on what type of work you do and your personal preferences, but if you’re looking for the most bang for your buck strictly on image quality the C70 value is about as good as it gets
The sensor is only half of it. How the camera uses that sensor matters. Many cars share the same engine, but how you tune that engine can produce double the power
So helpful, thank you!
정성스러운 영상 입니다 고맙읍니다
Awesome video!
Thank you.
Hi, this is handy, thanks for sharing!
If you can use davinci resolve for future color/iso tests.
Cheers.