Best Video Codec for Sony FX30 / FX3 (Apple ProRes RAW vs ProRes 422 vs HEVC vs XAVC S-I)
Вставка
- Опубліковано 6 тра 2023
- Sony Cinema Line FX30 (ILME-FX30)
- Firmware Version 2.0
- S-Gamut3.Cine/S-Log3
Recorder:
- Atomos Ninja V (Firmware 10.90.00)
FX30 lens:
- 3M-5A 500mm f/8 + VILTROX EF-E II 0.71x Speed Booster
- Sony E PZ 18-105 mm F4 G OSS APS-C (SELP18105G)
Noise reduction, editing & color grading in Adobe Premiere Pro CC 2023.
💾 Video in Apple ProRes 422 / 16G: drive.google.com/file/d/14tWD...
Full camera specifications (Official Website): electronics.sony.com/imaging/...
⚡ Leave any questions about camera and postproduction in the comments. Thanks for watching!
🔔 Link to subscribe to my main channel:
ua-cam.com/users/subscription_c...
Music: www.purple-planet.com
ENG
⚡ Instagram: / makeev.one
⚡ Flickr: www.flickr.com/photos/aleksey...
⚡ 500px: 500px.com/alekseymakeev
⚡ Reddit: / makeev_one
⚡ SF: www.studentfilmmakersforums.c...
⚡ Soundcloud: / cyberdroid
RU
⚡ Сайт: www.makeev.one
⚡ Telegram: t.me/russianfilmmaker
⚡ ВКонтакте: aleksey.makeev
⚡ Yandex Zen: dzen.ru/id/601fc39f9966a83871...
⚡ RuTube: rutube.ru/channel/24610261
#fx30 #raw #sonyalpha #noise - Фільми й анімація
Красиво.Замечательно снято.Молодец.
Hi, thanks for this great comparison video. Did you have any issue recording ProRes 422 HQ in the Atomos Ninja V? I've read online that there are issues trying to record non-ProRes Raw options with the Atomos and the FX30. On Atomos website as well it only lists ProRes Raw (and not the other ProRes flavors) for compatibility with the FX30.
Thank you in advance.
Супер здорово.Молодец
Супер фантастично.
Thanks for sharing my fx30 noise is shockingly bad
Interesting - I have found exactly the opposite- pleasantly surprised up through ISO 3200 (as high as I have ever shot) - using S-Log 3, overexposing a bit in camera, bringing exposure down in post and a touch of Resolve NR...
Hey great video. I have one question, does the A7SIII also shoot ProRes 422?
Thx! ProRes 422 available when recording on a Ninja Atomos V
Great comparison video, well done. To my understanding, the fx3 only crops in 10% in prores raw vs 20% crop for fx30. Would be interested to know if that difference and the larger sensor make Prores raw more usable on the fx3... Thanks!
It's more of an algorithm for getting raw data from the sensor, not cropping. The FX3 has a larger sensor area, but the number of pixels is much lower. 12.1 (4240 × 2832) versus 27 (6192 x 4128) on the FX30. Even the Lumix S5/S5M2/S5M2X has 25.28 (6000 x 4000). The FX30 uses the sensor for downscaling (like the Lumix S5). But the pixel-by-pixel RAW at 4.7K comes out with more cropping. That's why the pixel-per-pixel RAW with the Sony FX3 can use as much sensor area as possible. Plus the HDMI bandwidth allows for 4.1K data at 50 fps. But the 6.1K/50fps data from the FX30 sensor the HDMI protocol can no longer transfer at 12-bit. That's how Sony chose the highest possible resolution 4.7K of 50 fps for ProRes RAW on the FX30. The usual reviewers don't talk about this, because they are not engineers and can't understand the reasons for this decision.
Can it record ProRes HQ 422 without RAW but the whole sensor? Please advice…
@@alfnazr yes, in DCI 4K 23.97/24/25/29.97/50/59.97 fps
Great video man! So you're using the VILTROX EF-E II 0.71x Speed Booster on the Sony FX30? How's that working with the camera? I'm thinking of buying it but no where online dose it show that it's compatible with the FX30
Thank you! On the Sony FX30 it works as well as on other APS-C cameras. With some lenses there may be over-reflections from bright light sources (headlights, ceiling lights, sun). But similar on all speedbusters, even on Metabonce which is 3-4 times more expensive.
May I ask whether the prores422 in the test material was converted from prores raw or recorded externally with ninja v? On this basis, I found that prores 422 materials are sharper than xavc, but also introduce more color noise. In this way, can post-noise reduction get better results than xavc?
thanks
ProRes 422 captured from the camera and recorded in Atomos Ninja V
Yes, it really is sharper and the sensor noise is more correct because there is no AVC blocking.
Is shooting in ProRes raw really worth the extra steps in post production? Do you see that much of a difference?
I don't see much point in Prores RAW on the FX30. The picture is too noisy and the resolution is too low. The Lumix S5 allows RAW 5.9K from full frame is useful. But the 4.7K from the extra APS-C cropper on the FX30 is of very limited use.
which codec do you think is the best one?
As you can see from the test, there is no perfect one. It could be Prores RAW, but in this case the camera writes with a cropped image and only to the recorder. For 23.98, 25, 29.97, 30 fps you need Ninja V version 1, but for 50, 59.94 fps you need Ninja V+ (version 2). And that's a serious addition to the cost, because let's not forget about batteries and SSDs. 1TB is not even enough for 1 hour of ProRes RAW, let alone ProRes RAW HQ. I think that if you do not need very complex work on the post with VFX (chromokey for example), do not need night shooting with a lot of movement in the frame - then H265 is a reasonable choice. Yes it has flaws, but it's the best combination of size/capability. But this codec compresses not every frame, but only with skipping (Long GOP). So in scenes with complex motion and where in-camera noise modulation will be applied, artifacts may occur. But they will appear only at complicated post-processing and are not always critical. Also, the codec does not have frame rate 25. This is a long-standing problem of Sony. Maybe if videos like mine got to the top more often there would be more publicity about the problems.
Does prores 422 still crop?
Prores 422 work up to 59.9 fps mode (DCI 4K and UHD). Without ProRes RAW. And no cropping in these modes.
Кстати, хорошее исследование! Надеюсь @ImagingbySony обратят внимание и вместе с @sonyrussiacorp учтут в будущих моделях. Ну и тебе пришлют проверить результат