Jump to content

Andrew Reid

Administrators
  • Posts

    14,798
  • Joined

  • Last visited

Everything posted by Andrew Reid

  1. Absolutely, raw is still a benchmark for ultimate image quality. Let's be clear, I'm not denying that.   For me, now it's all about how close we can get to that beauty....but with more practical solutions like the GH4, A7S and D750.   I am over regular raw shooting now, as I ran out of space at the local aircraft hanger where I stored my hard drives.
  2. The real successor to the 7D for filmmakers is the C300 and Canon knows this.   This is why they don't need professional standard video on their DSLRs. It really is that simple. C300 sold incredibly well. Job done.   Canon thinks the soft, mediocre 1080p on the 7D is good enough for 90% of their intended user base.   All I can say to that is... hahha... now get back to work.   They are royally underestimated their customers. We demand more. Hobbyists and enthusiast with high knowledge especially absolutely realise how Canon are falling behind on performance terms vs the competition, for stills let alone video. Nikon has refreshed very recently across their entire line and Sony, Panasonic, Fuji, Olympus have all innovated with high end mirrorless stuff. Canon has done none of that.   For stills performance look at the aged 5D3 vs D810 or indeed the D750, which is cheaper than both. Or the Sony A7R for resolution and the Sony A7S for low light photography. Dynamic range is also a problem. Canon are 2 stops short. Then for video, ignoring the abject mess that is the DSLRs, the Canon C300 just does NOT compete on equal terms with the Sony FS7. It looks like 7 years older technology and when you see the 10bit 4K of the FS7 next to the 8bit 1080p of the C300, and the 180fps vs 30fps, you will see it in your bloody work as well. I'm not paying money for that kind of shortfall no matter how good the ergonomics and lenses are!!   What I find deeply odd, is Canon's complacency in the midst of all of this.... Vs the big guys like Sony and Nikon it is baffling enough. But then you add into the mix out of left field, Sigma(!?) making significantly better lenses than Canon, and their 35mm F1.4 outselling the Canon 35mm F2.0 IS so much that Canon had to give it a price drop... and of Samsung making more technologically cutting edge APS-C sensors... Samsung!? OF ALL PEOPLE!! I mean come on, wake up.   Canon's whole success has been built on leading the technology race. Best sensors, best cameras, best lenses.   Stills shooters STILL (if you excuse the terrible pun) have no high megapixel sensor from Canon. Why not? Nothing to beat the 36MP offered by Sony and Nikon.   Sony have the best sensor for low light photography and it does superb video. Where's Canon's answer? They don't have one!   Sony have a medium format sensor actually already in cameras right now, getting sold, and Canon could but doesn't.   Sony has mass market 24MP and 36MP sensors which are better performing than Canon's and not only that but they are giving Canon's biggest rival on sales, Nikon, an image quality advantage. It's all very confusing.   If Canon don't fix this, their good karma won't last and they will remember the dissenters like me in a few years wishing they'd listened.
  3. I love raw on the 5D Mark III. Magic Lantern worked a large miracle with it.   The thing is, the more the practical realities hit home, the more I am looking to make a few trade offs in image quality in order to get more manageable file sizes and a more reliable running camera.   I haven't yet had a single shoot where raw recording did not stop unexpectedly.   And I am missing shots because I can't record too much material due to space considerations. I recently did a shoot of a band in a music studio, 4 songs and each musician filmed separately laying down a track. Drums, guitar, bass, keyboard and vocals. I shot it with the A7S at 50Mbit/s, H.264 XAVC-S and ended up with 80GB of material after the 8 hour shoot was over. Can you imagine how much that would be in raw?   This is where pixel peeping fails and fails badly. It's actual of zero benefit when you have a situation as I described above. We can talk about compression artefacts and workflow until the cows come home, but if you have to start missing shots, or managing data whilst you should be shooting, it really interferes with the creative process.   Still a big fan of raw but... you've gotta pick the right tool for the job.
  4. Shots below are ISO 800, F1.4, same lens, same lighting.   In particular the colours are very similar.   After both were graded I couldn't tell the two apart.   The kicker...   For this clip, a few seconds long... D750 weighed in at 10MB and 5D3 raw file at 500MB. What's more, the D750 grade took me about 10 seconds to match the clip from the 5D Mark III. Quick change of the gamma curve in Resolve, and a saturation boost, then done. The auto white balance did a superb job in-camera. The 5D Mark III raw I can sit and grade for hours. Obviously you have more control, more freedom, but at what cost?   I just want the image... and the D750 appears to give me that.   S-LOG2 on the A7S does too... but it's a bit of a hassle.   5D Mark III 14bit uncompressed raw     Nikon D750 compressed H.264 at 24Mbit/s (24p)     By the way, you can tell the difference. The smooth gradation and transition in the mirror of low contrast shades looks blockier on the Nikon shot and the noise grain is finer on the 5D Mark III.   However this wasn't shot via uncompressed HDMI and the difference isn't large enough in my opinion to suffer the file sizes... of raw OR ProRes for that matter, unless you have a lot of fast camera movement, tricky lights and motion blur to contend with then the Atomos Ninja Star on the back of a D750 will really help.
  5. Presumably pairing it with a Ninja Blade would avoid that problem. Because the 20 minute limit is for the internal recording.
  6. Let's call it a purple vignette. Purple fringing is something else, to do with lenses.   The purple vignette I've not had yet, how long was the camera running for when it happened?
  7. Downloaded the original 4K file and it's looking good.     More from Philip here - http://philipbloom.net/2014/09/29/gopro4/
  8. Already have it in Canon mount. Damned annoying.   Situation would be a lot more simple if Canon would just give us a body that was actually better than the last one for video. After 3 years you'd kind of expect them to do that.
  9.   I have a lot of Canon IS lenses that I can't use on the Nikon, and can't afford to replace all of these with Nikon VR lenses. That's what I meant in the article when I say I will miss Canon's stabilisation as opposed to no-stablisation on my Nikon glass.
  10. Depends more on encoding quality than bitrate.   Compare Fuji's codec to the C100. On the Fuji X-T1 it is 36Mbit/s and on C100 it is 24. Nobody will find the Fuji codec holds up in the same way.   Noise and blocking in shadows is introduced by a bad encoder. Wise maths with 24Mbit/s is better than dumb maths at 48Mbit/s!
  11.   It's flat but strongly saturated. On this shoot I didn't have contrast dialled down all the way. You can go even flatter if you want to. I've yet to see if that's an advantage or not but with the lows being so cleanly rendered by the new Nikon codec and the sensor output being so smooth, it really does have a chance to break the 12-13 stop mark for dynamic range in video mode.
  12. No filtration, it was slow-mo so needed the higher shutter speed anyway and wanted to shoot it pure to gauge what the camera was doing.
  13. So custom white balance and turning saturation all the way down in camera might be a solution for the A7S then?   It remains to be seen how the D750 handles it but the codec is just so clean... colour is lovely from the new flat profile once you apply your own curve in post. I also like to lift the blacks and when I do that there's no noise or banding. It's very nice.
  14. Filmmakers have been waiting a long time for a top-performing Nikon full frame FX DSLR for video. Is the D750 the one we've been waiting for? Read the full article here
  15. Good video to show the problem with.   I've always had similar issues with bright blue lights on Sony cameras. FS700 has it too and that's $8k. It needs to be fixed.
  16. Tried it, yeah works through the viewfinder but is kaput in live-view.   Looks like a bug!
  17. I think this thread should now be left to the input of other users. If there's enough interest from them to see it gravitate to the top of the forum daily then it should stay, and if not then it should sink.   John, please refrain from bumping it up on a daily basis.
  18.   I'm afraid I agree with my readers on this John. What seems to be going on here isn't really fair. The selling "rules" are not a set of laws to be enforced by mods, they're meant as a general guideline to staying safe and as a disclaimer. Also your bad relationship with some of the other mods isn't sustainable. I'm afraid from now on you will no longer have moderator rights on the forum John. Sorry.   I am sure you can appreciate my position here. I feel you are abusing your power as a moderator to sell your stuff and censor competitors and this is just not acceptable. Way too many complaints for you to stay as a mod.
  19. Nope. Only difference is it doesn't hunt when it snaps onto focus, but it snaps on a lot slower than the best contrast detect AF and is hopeless in low light.   What are you hoping to use it for? If it is casual holiday snap shot video time, then fine it works well. If it is filmmaking, then no, you need MF. And if it is stills forget it, just look through the optical viewfinder instead.
  20. What on earth is this thread but a total waste of everybody's time. Pathetic LAfilm.
  21.   Does it? Big shortfall in image quality for video to the Nikon D5300, not to mention the others.   And dual pixel AF is too slow for stills, too unreliable for filmmaking.   But apart from that... yeah... wow... amazing. etc.
  22. I have a D750 to test so will put it through its paces on the blog this weekend.   Looks good so far. Flat picture profile offers pretty impressive dynamic range and there's very little weirdness coming from the sensor. At least as clean as D5300 but seems less soft.
  23. Yes some side to side test needs to be done to see which is better.   That's why you need a site like EOSHD.   Unfortunately nobody at Nikon have bothered sending me the cameras. From my files I shot in Cologne and at a shop, they do look very similar but the D750 looks a bit better in low light.
×
×
  • Create New...