Can you tell more about "NVIDIA limiting encoders "?

I found a way to remove that limit, no problems for about 3 months of use https://github.com/keylase/nvidia-patch

That's totally incorrect @Vitaliy_Kiselev

Vegas uses multiple cores and gpu rendering for timeline performance as well or better than any other nle these days. Legacy and 3rd party fx may often be cpu single core only plugins that bottleneck vegas pro, but the base engine can use 32 or even 64 threads, with gpu acceleration to give smooth timeline playback using 4K-60p h.264 with fx applied. Even with 4K-30p h.265 hlg footage timeline playback is perfectly smooth on my machine. My issue is rendering where another guys TR1950x Vega 10 frontier edition machine shows the fastest rendering tests ever seen in vegas using Magix vce encoding (faster than an i9-9900 gtx2080ti machine using nvenc encoding even). My machine takes over twice as long with a supposedly better gpu thus my question.

I have some sort of wattman throttling going on and can't sort out where it's coming from.

Cheers, Pete

@c3hammer

I just built up a tr1950x, asrock x399, radeon 7, 32gig ddr4 2400 box and I'm getting just mediocre performance from the MSI radeon 7. Everything else shows good results in the few tests I've done except the gpu. Rendering the 'red car test' in Vegas Pro 16 is about double the times that others are getting with lesser AMD setups

Vegas can have no use for powerful GPU in this test. As for better results - they can have better CPU (for Vegas it can be fewer cores that are faster!_ and memory that is much faster than 2400. Most of Vegas code and routines are very ancient made for 1-2 core CPUs and with lack of parallelism in some parts.

I suggest to check in Resolve after reading how to do proper config and such.

Would this theoretically be the best thing I could buy for an external GPU mac setup on FCPX?

I need new graphics card, and NVIDIA limiting encoders was a terrible discovery for me.

Can you tell more about "NVIDIA limiting encoders "?

Navi will be low end cards, much slower. And will be available only around one year from now.

More tests

In DaVinci Resolve we have an extensive project of more than an hour to eventually export image material as a h264 file. In it, the Radeon VII is the fastest video card, but the RX Vega 64 was also very fast and could halve the total render time of the RTX 2080.

https://nl.hardware.info/reviews/9051/19/amd-radeon-vii-review-nieuwe-kans-voor-vega-op-7-nanometer-extra-davinci-resolve-blender-gpgpu-en-overklokken

Same strange numbers repeat.

https://tweakers.net/reviews/6842/17/amd-radeon-vii-videokaart-vechten-tegen-de-rtx-bierkaai-gpu-versnelling.html

We put together a series of tests for both Premiere Pro and Davinci Resolve exporting 8K and 4K footage into HD, with and without GPU-accelerated effects added. It's not worth showing the results, though, because the difference between the various cards was negligible. Using no GPU acceleration noticeably slowed down the exports, but the difference between our older 980 Ti and the Radeon VII was only a few seconds.

https://www.engadget.com/2019/02/07/amd-radeon-vii-review-video-4k-benchmarks/

Unboxing videos are up

14 phases using, as it seems, IR power stages, plus few other smaller phases for stuff.

On board we see 4 phases not soldered.

All tantalum capacitors (looks like black flat things not resembling normal capacitors).

So, no ray tracing to do Hollywood level 3D effects, and something closer to, or less than, a rtx 2080TI, or less than better Pro cards, at 8k video editing?

I'm also interest in knowing this question, on how will it behave in Adobe, they claim a 30% increase but that's weird. The difference between OLD GTX like 960 and 1080TI are unnoticeable unless you render/scrub huge 4k massive files, don really know how the new RTX card are performing but, those 30% are at least intriguing.

I really hope Puget Systems gives this card a chance in their suite of tests. I'm very interested to see how it manages in Adobe CC applications (outside of AMD's 'benchmark data' there).