[Angelo Lorenzo]

I believe the mac pros use dual channel memory. Basically each pair of RAM slots will be married to each other (I would assume they have some kind of color coding) You should, to maximize the speed of your RAM, make sure each stick of RAM is paired. All memory, ideally, should be the same clock speed, and each pair should be the same capacity (you can mix pairs of 4’s and 8’s, but not so much three 4’s and one 8 for instance)

I’m using Corsair in my system, although Kingston is a good choice as well. Crucial is reputable, but I’ve not heard of OWC.

[Angelo Lorenzo]

Well your issue isn’t exactly like the OP. The OP sounds like they’re using a mix of Premiere and AE which, for optimal performance needs a cuda enabled card (for premiere and colorista, the effect he uses quite a lot) and RAM and processor for AE.

Where are you benchmarking? Render time purely in Premiere/Media Encoder? There are other bottlenecks to consider including disk speed (are you using an internal, firewire or eSATA drive, and so on).

It’s a question with a different answer for everyone.

[Angelo Lorenzo]

I’ve not used Davinci personally (I use PC predominately and I’m working with 2K-5K RED footage, and Davinci Lite has a 1080p limit) but I do know the roundtripping uses the export XML feature. This is the same XML you’d use to roundtrip to FCP 7 which leads me to believe you may have some issues if you use more exotic transitions that don’t convert well. Another option is to render the Premiere file and use an EDL for Davinci to show all your cuts, and then you just keyframe color grades over fade transitions, etc. It really depends on how much you’re doing in Premiere before it touches Davinci.

Make sure you’re using the most up-to-date PPro, there were some issues with PPro using FCP’s XML version 4 while Davinci only likes version 5. This is now corrected.

When it comes to color grading, learn how the lift, gamma, and gain controls work, and learn how to read scopes.

https://www.amazon.com/Color-Correction-Handbook-Professional-Techniques/dp/0321713117/ref=sr_1_2?ie=UTF8&qid=1328679620&sr=8-2 is an amazing book.

https://www.taoofcolor.com has a good but kind of pricy online course. They also have a great podcast with the author of the above book https://www.taoofcolor.com/227/alexis-van-hurkman-interview-part-1/

[Angelo Lorenzo]

The Mercury Engine is the playback technology, when you don’t have a supported CUDA card it just falls back onto “software” mode.

Before I start, don’t bother with After Effects in this situation. It isn’t made for non-linear editing (not with ease) and isn’t designed to play footage in real time.

Sadly, from experience, I can’t recommend dealing with hi-def footage on a dual core machine, at least not in h.264. Your best bet is an editing codec. Avid supplies their codec, DNxHD, for Windows and Mac for free https://avid.custkb.com/avid/app/selfservice/search.jsp?DocId=263545 and you should see some better performance once your footage is converted. When installed it’ll be available as a quicktime format in your export dialogs and Adobe Media Encoder.

[Angelo Lorenzo]

Probably best to pick up a book like https://www.amazon.com/Mixing-Audio-Second-Concepts-Practices/dp/0240522222/ref=pd_sim_b_2 because there is so much mixed information online.

https://audio.tutsplus.com/ isn’t a bad resource either.

Generally, when looking at a dialog track, I do the following:
— Look at the general structure. Is there a lot of noise or a little noise, are there steep peaks in volume or does the speaker use even diction and volume, and so on.

— If the audio is really chewed up (audio is close to the noise floor) I’ll normalize the peaks to zero so I have some working room for noise removal, compression and re-normalization.

–Depending on how heavy you process, do noise reduction, eq, multiband compression, and then reverb. There are no hard and fast rules but it’s a signal chain I like. I prefer to do everything that affects volume heavily before compression… you’ll also find people who like to compress the signal during recording (mostly in a controlled studio setting).

[Angelo Lorenzo]

Go to Edit > Preferences > Audio Hardware and make sure you have an output set, then go to Audio Channel Mapping to make sure left and right are mapped to your output device.

This can happen often if you are using Audition with usb speakers or headphones. Your computer sees usb audio devices separately from your internal speakers or normal headphone jack. Once the usb device is unplugged, Audition loses it’s output device and mutes itself.

[Angelo Lorenzo]

The rule of thumb for multicore processing you’ll hear is two gigs of RAM per core used.

So lets take a look:
AE & Premiere share 12 gigs of RAM according to your settings. Usually, Adobe programs split memory management evenly; if you turn multicore processing on then try not to have Premiere open or plan on optimizing your settings for 6 gigs of RAM.

Try running only three processor threads (3*2=6 gigs of RAM) and check performance.

[Angelo Lorenzo]

I’m not sure I understand the question.

You know how to:
Edit the L/R balance of a track in the timeline using the mixer or keyframes

I assume you want to:
Change the balance of a single clip so it is always applied like that as you add it more places in the timeline, or other timelines in the same project.

If that is the case, then the answer is not really. There is very little L/R control when it comes to interpreting footage (right click your clip in the bin, go to Modify > Interpret Audio). In multichannel audio you can move what channel goes to what speaker (with no percentage mixing), and mono tracks can either come in as stereo hard panned to one speaker or the other, or as cloned stereo which feeds both speakers with the same track.

[Angelo Lorenzo]

I don’t know, what are you doing?

There is no one fantastic perfect workflow. If you’re editing a 60 minute documentary with a few titles and a ton of lower thirds, you can keep everything in Premiere, color grade in Premiere, and dynamic link in the motion graphics.

If you’re editing a feature-length film you may want to do all your editing in Premiere and then bring the full film (or each sequence in) as layered comps after edit lock so you can do your visual effects, camera smoothing, and heavy color correction and grading with multiple masks and layers.

[Angelo Lorenzo]

Well since you’re asking in the After Effects forum, I assuming you’re doing your grading and conform in AE, and just the cutting in FCP. Consider grabbing Automatic Duck. It’s now free and, if I recall correctly, it will take a FCP timeline with prores and replace them with the R3D files when creating the After Effects project (basically putting the R3D files online for you).

But to answer your question, you can add effects to R3D files in After Effects… I have a full timeline of Epic footage right in front of me.

https://www.amazon.com/Color-Correction-Handbook-Professional-Techniques/dp/0321713117/ref=sr_1_2?ie=UTF8&qid=1328679620&sr=8-2 is worth buying and below is some info on “day for night” distilled from the book.

When shooting day for night make sure the DP is looking after the following:

– Keep the depth of field shallow: a visual cue that confirms a low amount of light

– Frame the skies out out of exterior scenes or consider using graded ND filters

– Avoid objects with high saturation. Low light situations usually result in muted colors in the shadows.

– Throw as much of the background into shadow as possible. Deep, evenly lit scenes don’t transition easily to day for night.

– Under expose by 1-2 stops. This helps mute color and contrast but maintains healthy midtones and highlights.

For color grading, day for night usually means “relighting” the scene by drawing multiple masks to reenforce shadows or pools of light with lift and gamma controls. While cooling the color of a scene might seem like a good choice, make sure it works in context. If a scene is lit by sodium street lights, incandescent bulbs, or candles then you go more yellow/orange.