I haven't scrutinized the material but after applying the Resolve Arri to 709 LUT it looked good on the scope, not quite down to zero and peaking between 650 to 690ish millivolts. This was a multicamera stage shoot of musical performances with fancy lighting and pyro what I've seen looked pretty good and reasonably matched. These are big stars so I can't tell how it was crewed but the lighting and staging is strong. We've also recently done comedy standup specials and I did hear there was some concern about the comics flesh tone drifting or losing some saturation as the shot played. I haven't seen what they flagged yet. I'll know more in a week or so once offline is done and I get to color correcting.
I can tell you I've done 3 or so comedy standup specials with the Arri Alexa and the 6 cameras weren't that hard to match up and they held their exposure well for the stage comedy standup shows.
I got so excited when I saw that I could get better than 3 times real time transcoding in resolve with the 29.97P ProRes HQ Raw to DNX 220 I really want to know if I can do better than that if I get the source material on a Graid that has esata connection and I transcode through the Resolve system's fibre card to one of my online drives on SanMP. I hope a get a chance to compare that with an Avid ama transcode. We don't have Ver. 7 yet so I won't be able to add a LUT for that test. Anybody know if adding a LUT in Avid Ver. 7 and above add much to transcode times to bake the LUT in. I know you don't have to but it helps down the line when transcoding to offline resolution. These projects are being done out of house using single media drives to we have go down to 10:1 to get the multigroups to playback without too many hickups.
---In Avid-L2@yahoogroups.com, <Greg@...> wrote :
On Dec 10, 2014, at 22:26, John Moore bigfish@... [Avid-L2] <Avid-L2@yahoogroups.com> wrote:
Today I had ProResHQ Raw files logC from an Arri Amira camera. On our 12 core MacPro circa 2012 2.66GHz with a nVidia GTX-680 hooked firewire800 to a Graid drive with the source files and rendering Avid media with an Arri to 709 LUT applied it took 4.5 hours to render a 6 camera stage shoot. According to the ticker on Resolve it was rendering out at about 35 to 40 fps flickering up and down on the rate. After my transcode was done I set up the same render only this time the source was the same Graid hooked with firewire 800 but I changed the destination to an external 8 drive raid enclosure connected by Esata. This render was clocking at 96 fps and the estimated time was right around 2 hours, more than twice faster. Of course going to a different drive raid with Esata made the difference. I'm curious if my source drive was also Esata or something else would I see even faster rendering? I've read on the Avid L2 how sometimes being on different busses in the computer makes a difference too. This mac pro Esata is on a pcie card and it has two outputs. I'm wondering if I had the Graid source also connected to the 2nd Esata connection on the same card as the external Raid if that would hinder the throughput as the two connections would certainly be on the same buss if they are connected to the same card?Given I was able to get better than 3 times real time going to the external drive and the AE told me the ama linking and transcoding usually went around real time I'm curious what others have experienced. I know Avid is optimized for transcodes so does anybody have a ball park render estimate of how fast Avid would take Arri Amira ProResHQ logC raw files and transcode to dnx220? Would it be anywhere near 3 times real time. I had to get the job done and didn't have time to experiment with the Avid ama transcode times.
Posted by: bigfish@pacbell.net
| Reply via web post | • | Reply to sender | • | Reply to group | • | Start a New Topic | • | Messages in this topic (3) |
No comments:
Post a Comment