Aaron Star
Forum Replies Created
-
Same thing. Learn how to key frame the Sony color corrector. Both examples are easy.
-
Is your GPU rendering turned on?
-
Yes. It looks like some simple layering of text over graphic layers.
-
The full frame sensor is great for that 35mm look for still shots.
The DX sensor is actually the same size as 35MM motion picture film. The gate size in motion picture film is smaller than 35mm still due to orientation, perfs, and allowing space for audio tracks. So for shooting video, the DX is closest to the 35mm film look.
MP Film Gate = 22mm x 16mm
DX Sensor = 24mm x 16mm
From what I have read all but the latest Nikon full frame dslr have moiré video problems. I think Nikon has worked out the issues with the 750 and 800E.
Magic lantern on the Canon side, seem to have less problems, and the 5D Mark 2 and new cameras can shoot RAW video for a less video look.
I have a full set of Nikon primes from my F3 days, and so lean towards Nikon myself.
-
Is the footage from the EX1r XDCAM format? If so do a test to Sony .mxf > XDCAM that equals your project.
Then try handbrake on the output file, use the normal profile in HB.
“3 tracks Logos – track effect 2-d shadow”
drop shadows can load the CPU up, a properly functioning GPU assist will speed these calculations. The more advanced the compositing, the long it will take a CPU to render the frame and then pass it to the encoder which also takes time to compress. Grazie is referring to altering your edit so that it reduces complexity of the render.
“3 Tracks – A – B roll – color corrected – track sharpen”
These are fine, but much of what I said above also applies to these plugins. Most of what I see in your system configuration is your Nvidia card not properly performing the GPU assist, and holding you back. The CPU+GPU is used to render the timeline effects, then that frame is passed to the Encoder. The Encoder can utilize the GPU for certain calculations that speed up the encoder. Swapping out your NVidia card for AMD 7970-GHz or even a R9-270X, and you would see a bump in speed for more complex track compositing.
Since your camera shoots XDCAM, not having OpenCL/GPU acceleration enabled, you are missing out on XDCAM playback optimizations in Vegas.
In order of calculation performance
R9-270x (new)
HD7970-ghz (ebay)
R9-290X (new/ebay)At some point your system/CPU will be holding you back. GPU is not everything, its more like a 4×4 truck or hybrid car. The GPU assists with things it can do faster.
Upgrading the GPU may require a new power supply.
-
15.7.1 on my 5770 here, and I find it stable.
The 7790 is very close in GPU config to the 5770. A 7970-GHz or 290X would probably better match for your 4th gen i7. You can see AMDs little shell game with model numbers. Just a thought.
There is a nice “AMD system monitor” that gives a good GPU utilization read out.
-
Your system is just shy of being able to use the GPU, so keep that off, and do not set GPU in your render templates.
Your project settings should be:
Size/resolution: 640×352
Pixel ratio: 1.0 square
Field order: progressive
Pixel Format: 8-bitFrame Rate: 29.97 or 30fps depending how your source was recorded. “MediaInfo” application can help to determine this exactly.
For rendering try Sony AVC, and create custom internet profile that matches your source file format. For bit rate choose about 2-5mbs.
Scaling your video to some large format like 4k or 1080p may cause playback issues in BEST/FULL mode. Making sure your project settings match your source media, and try playback with Preview, or Good mode in the preview quality selector.
Trial mode on Vegas should not hold you back. Jerky playback with your project matching your source, there may be something else limiting playback with your system.
-
$2000 will buy either of the cameras I mentioned above. If you want a DSLR that shoots video, and then can switch modes and shoot stills, a video camera will not work.
“Video cameras” are setup ergonomically for the things you do most in video shooting. Video cameras will generally have better sound recording built in too.
DSLRs are setup with controls that make most sense for shooting stills. Video controls on DLSRs are generally treated as a second cousin to the Still mode. Audio on DSLRs is generally pretty bad and so people use a second recorder to make up for this.
So it depends on which you plan on doing more of Video or stills on your trips.
DSLR for less than $2000 new, you could go ebay for different used pro rigs, I would look at the Panasonic GH4, Sony a7, or the devices above in descending order. Any of the DSLRs will shoot auto modes for beginners. The GH4 has functionality to grow into. The simpler cameras above have less pro features but are smaller, and less prone to theft. Nikon vs Canon vs Sony Vs Panasonic, you pretty much need to pick your religion. Lens mounts are generally the difference between them, but there are converters to use a lens from a different maker. The advanced modes and auto focus generally go out the window when converters are deployed.
Video Camera less than $2000 – Black Magic pocket cinema series with a lot of storage and a good lens. This is an advanced to intermediate DSLR type video camera. It will take some quick study, but your video out the other end of posting it will be much better than a standard DSLR video. Still would need to be grabbed from raw video frames via short video bursts.
Sony FDR-AX100 is pretty much your typical Sony “handycam” with todays 4k video specifications. The lens and sensor are nice, as well as the 8-bit 4k video from it. I believe it even has a 5 channel surround mic on top.
One other thing to keep in mind is, some places will not allow a full on Pro DSLR to be brought inside. So a tourist form factor like the Nikon A or Sony AX-100 are less threating and can often times get pass security checks at clubs and museums.
Looking back to the “day” my Panasonic DVX100 was rejected by security in front of Mans Chinese Theater in Hollywood as being to professional to shoot with. Of course we were shooting a short film at the time, posing as tourists.
More recently my old Nikon D5100 was rejected by club security in Las Vegas only a month ago.
-
If you render your sequence to EXR, TIF, or PNG, does the problem still exist?
Are you rendering h264 out of Blender, then uploading to YouTube?
What it looks like to me is a shift in Color Space that may also have different Gamma.
Rendering from blender to h.264 would maintain correct levels directly to YouTube. However, rendering to JPEG, then to h.264 is introducing extra conversion layers. Understanding the conversions and how to adjust for them is more likely the problem with that workflow.
from the Blender Manual:
“If you are going to do post-processing on your movie, it is best to use a frame set rendered as OpenEXR images; if you only want one file, then choose AVI Raw. While AVI Raw is huge, it preserves the exact quality of output for post-processing. After post-processing (compositing and/or sequencing), you can compress it down. You don’t want to post-process a compressed file, because the compression artifacts might throw off what you are trying to accomplish with the post-processing.” -
I would do full tests both ways at your laptop native resolution and at 720P. The drop in quality might be from your laptop display scaling the 720p render. YouTube is expecting conforming video, or you suffer the wrath of the Scaling. Render at Best Full will help with scaling the pixels to 720p, but adds time to the render.
Since your upload is bandwidth is so low, you may want to drop your capture to 30FPS, and render 30FPS at like 7-8Mb. This will reduce your upload size greatly.
1080p is not a waste in the era of tablets and phones with 4K displays on them. 1080p is just rough to upload at such low bandwidths.