Hillary Knox
Forum Replies Created
-
Thanks Walter. If I actually do get to conclusively test this, I’ll post results.
That’s some interesting info. I never thought about how read speed would interact with writing to the cache. I didn’t realize it’s caching *each* layer * individually in addition to caching *all* layers.
-
What is a better scenario for AE cache location…
1) in a subfolder on the OS/Apps drive, which is 2 SSDs in RAID 0
Or
2) on a 12-hard drive RAID 5 SAS which also contains source media (and is somewhat faster than the SSD RAID)
-
So wait, you can’t buy a new license of Lustre 2013 Windows, can you? Or are they just updating in order to not piss off the small subset of folks that happened to buy Lustre on Windows 7 years ago?
-
Hell yes I fully support that request. I mean, if they really want to do it up right, they’d completely open up custom mapping – if, for no other reason, it would save so much internet debate about the “correct” way to map 8 “log” parameters to 6 physical controllers.
-
Before you kill yourself & spend a fortune on anything, if I were in your position I’d definitely at least have at look at some consumer-grade daylight color-temp LEDs.
As far as color, I don’t know what the tolerances are for the consumer-grade LED stuff – and I don’t know of anyone taking a light meter to the task to either confirm or deny whether this is a good idea or not, but I doubt it’s that far off. I do know that “batches” of LEDs can vary from one manufacturing batch to the next, and that there is a visible-to-me tolerance between what is an acceptable variance and what is not. My bet it that if you got fixtures from a single batch of LEDs, you wouldn’t notice the difference, but don’t plan a facility around that advice.
-
Thanks for the replies.
Short answer: thanks. it totally worked for what I need
Long answer: I did 4 versions (2 of them might be superfluous): 2 versions using Time Remap only, 2 versions using Time Remap plus Timewarp. I can *maybe* see a micro-difference with Timewarp in addition to Remap, but it could just be psychological.
I did one batch with the Frame Blending set to Pixel Motion & precomped that prior to doing the Time Remap/Timewarp just to see if it would make any difference. Again, maybe micro difference, maybe not. The only reason I did that was because in a Kronos tutorial I saw, that was the first step before they Kronos-ed the footage & I thought it might be worth a try. Don’t ask me what it actually does.
Again, thanks for the help.
-
@andrew smith If I were you I’d download the trial FCP version & see if you think it’ll do what you need. The end result will/should be the same as going through Neat Video in AE. Well, maybe…you’re not really working in the same color space in FCP & AE, but I’m not sure what kind of difference that would make. I’m sure someone more expert than myself could weigh in on this.
-
Yeah…I’m probably gonna get killed for saying this out loud, but I find it, shall we say, difficult at best, to get a great-looking graded image from 5d/7d/etc footage without using NR on some level. It makes me sad to hear that Resolve’s NR isn’t quite cutting it for a lot of people. I like the results I can get from Neat Video, but it is SUCH a speed-compromised process either running everything through AE before grading it, or especially, grading in AE just to be able to use Neat Video.
I’ve played with NR in Smoke, and have seen it done in Nucoda & Lustre – all pretty impressive stuff sans the AE frustration. I assume Baselight would be equivalent or better. I’d be curious to see a quality comparison between AE/Neat Video and the big boys with 5d or other characteristically noisy footage.
-
As for masking in AE, I’ve never done it, but it should be pretty straightforward. I imagine that you’d duplicate the layer, do the appropriate masking/tracking/animation, do the NR on the masked part, and move on with your life…Not quite sure how to do it in FCP, but I imagine it’s possible.
-
Personally, I have tried Neat Video with both FCP and AE. I bought the plugin for AE because it seemed a little less sluggish in AE, plus, as I mentioned before, when I do grading in AE (yeah, I know…) I like to have NR as an interactive part of the process without having an extra render generation in there.
Quality-wise, I think the result is pretty great…BUT remember to bring your patience. It’s not what you want to necessarily be doing with a tight deadline. Forget the words “real-time playback” because they just don’t apply in that universe – I mean, that’s AE for you, but even so… Just to maintain my sanity, I end up turning off the NR while I’m grading, and then turning it back on & checking it right before I render. Even though I think it’s more sluggish in FCP, being able to render the timeline might be worth it to you. Also, in either app, the final render times can really start to add up when using (I assume) any NR. I’ve also tried Magic Bullet Denoiser, which I think does a decent job when it works, but it’s been a little too crashy and render-corrupty for my taste.
Ideally, Neat Video wants to work on individual shots as opposed to doing one massive de-noise on the whole show. I suppose you could make it work that way, but it won’t give you the best results. So, what you end up with (in AE) is each shot on an individual layer, which can add up to several hundred layers – which can be off-putting for some people.
I would say download the trial version for both platforms, play around with them, and that will give you a good idea of what you’ll get.