jan1

I finally did find a difference with the Quadro card. It's actually not so much with the heavy duty benchmark tests, but with just plain playback of footage. I think the heavy duty benchmark tests get bogged down in the processing, that copy bandwidth is not the bottle neck. But taking just the plain 8K 60fps RED clip (the shark), I get a distinct 10fps advantage in playback rate on the Quadro card compared to the 2080. I should caveat that this with the new CUDA RED debayer, so the GPU does the heavy lifting for the RED footage, which presumably adds extra I/O overhead to the GPU. Putting this clip on an 8K 60fps timeline, and configuring output via AJA to UHD 60fps, I get the following results: (without I/O) RTX 6000: 53-60fps for RED clip as is RTX 6000: 54-60fps for RED transcoded into 8K 60fps ProRes HQ 2080TI: 42-43fps for RED clip as is 2080TI: 36-45fps for RED clip transocded (with AJA I/O) RTX 6000 - same numbers 2080TI: 46-48fps 2080TI: 25fps for RED clip transcoded, this is a curious result, not sure yet So the Quadro card can get a leg up if you have super high res / super fast footage and are doing RED decode.

I've created a somewhat similar test on Resolve, so I can compare them. They're not identical matches as it's not easy to set TNR and DeNoise to the same parameters, same as with blur. But they're principally very similar. I've now put the 2080TI back into the system, so I can run Mistika on either one for easier comparison. Here are the results I'm seeing on the Quadro RTX 6000 Mistika (with AJA I/O) TNR 6.8fps peak, then 2.6fps after VRAM fills up 8 Node w/ blur at 6.4fps 12 Node w/o blur at 4.8 fps Resolve (with DeckLink I/O): TNR 10fps steady 8 Node w/ blur at 21fps modified 8 Node w/ blur at 13fps (In the first Resolve test I have a layer mixer with 8 inputs. I believe Resolve realizes it's the same source media, so for the modified test I created 8 separate copies of the clip and layered them on the timline with overlay mode in the timeline, not a layer mixer to approximate what we see in Mistika) After forcing both Resolve and Mistika to use the RTX 2080TI instead of the Quadro RTX 6000 I did not find any noticeable differences in render speeds. I verified utilization via task manager to make sure the right GPU would be used. Where does that leave me? With a few questions.... 1) It seems that while the full-duplex bus management is in theory an advantage, it doesn't seem to be a frequent bottleneck. At least I have not been able to construct a test that can demonstrate it. 2) While the two tests are an approximate pair due to tool differences, it does appear Resolve has much more efficient playback as it outperformed Mistika by 2 to 3x. Of course this is not a very exhaustive test, as I was looking for one specific aspect, which hasn't born out yet. At present the extra cost of the Quadro card other than long-term stability doesn't bear out in performance differences. To be continued....

I've created a slightly different test timeline in light of the full-duplex copy. 1st media clip: https://arriwebgate.com/directlink/b043cddb9c4658e0/1684269 (Alexa Mini LF RAW F003C106) 2nd media clip: https://arriwebgate.com/directlink/77689a9b6ef6ddc1/995144 (Alexa Mini Arri RAW M002C006) both from https://www.arri.com/en/learn-help/learn-help-camera-system/camera-sample-footage The timeline has three stacks. They use UHD or bigger footage and ACES color. In the first stack I did a combination of keyed color and blur (like Cristobal's example) and also added temporal NR. Then stacked these pairs three times. The other two stacks have groups of the same footage to create additional textures. They're then combined via Comp3D and overlay. The second stack uses a combination of key, track and blur. The last one stacks higher but omits the blur which is more CPU focused I believe. The first stack tests more GPU compute and also memory. It starts using more VRAM until it runs out and then swaps. Interestingly enough once it starts swapping it never releases the GPU memory. You have to exist and restart Mistika to get the better performance back. 1st stack: 6.8fps until VRAM fills up, then 2.6fps 2nd stack: 6.4fps 3rd stack: 4.8fps On the second and third stack I start seeing up to 8% copy utilization in task manager which is what I was after for testing bus management. This is on the system with the Quadro RTX 6000. Later today I'll install both the 2080 TI and Quadro and compare the two against each other. Obviously an extreme scenario to stress test a system. MistikaTest_1.env

Sorted the audio issue. Two apps competing for AJA. I can playback at 80 layers in RT w/ audio.

Here are my results: Real-time playback with all 80 layers, no loss of sync indicated. I added more layers to 100, still real-time. GPU load is 100%. But I can't hear audio. Something is wrong in the routing. Can't hear audio even on the clip on its own. System: Puget System Build, i9 16 core CPU 3.1GHz, boost to 4GHz 128 GB RAM 1 Quadro RTX 6000 GPU

Thanks @Javier Moreno. I will put the 2080TI back in this weekend and do some more extensive testing. I have one 8K 60fps RED clip that has been good to test with, and I'll find some other UHD clips form Arri as well.

Fantastic. Just what I was looking for as I'm testing a different GPU for my system. One thing I'm specifically looking for in this test - a way to see if Mistika can take advantage of the full duplex loading of the textures. Do you know if this test will stress this? Also second question - if I have multiple GPUs in the system, is there a way of telling Mistika which GPU to use? I looked in mConfig and didn't see anything. I recall a support article mentioning this, but can't find it right now. I'm thinking of leaving one 2080TI in side-by-side to the Quadro RTX 6000 so it's easier to compare the two. Same question, can I tell Mistika to use one for the UI and the other one for all processing? I know Boutique is not multi-GPU enabled, which is fine. But can it be told which GPU to use if multiple are present?

Received my Quadro card yesterday and will run some performance comparisons. Does SGO or someone have a nice Mistika project they use for GPU stress testing? Something similar to the Resolve Standard Candle Benchmark? Anything specific I should be looking for comparing the Quadro vs. dual 2080 TI? I have downloaded various RED and Arri footage, some 8K 60fps. So I'll put those on the timeline and see what playback performance is as is and with some grades.

Of course. But that will be an issue with any new system, regardless of OS. When I first installed Mistika on the Mac it took a very long time to work out lots of kinks. You also have a big learning curve on Mistika ahead of you. I've done a dozen or so client projects now, and I'm getting faster and using more of the system now. I think you have to be realistic about the productivity curve regardless of choice. I get the aversion to Windows, I'd rather avoid it too. But at this point, despite the fresh off the line MacPro, I don't trust Apple enough anymore to rely on them 100% and I also find their ecosystem to constraining. But YMMV.

I should also say it's been improving since I built the system. Yesterday with the help of Wacom we figured out which old driver is stable for touch on the Wacom - the answer is the June 19 version, not the latest. Other things are also slowly getting there. And I have a lot of workflows on this system, so there's much to tune. On a simpler system it should be less of an issue.

It's a bit of a mixed bag. No issues while running Mistika. The stability issues are generally elsewhere.

That might be something for the feature request list then. While the Ctrl+PgUp solution is handy, making scrolling of this more easy for a variety of tablets/mice would be very helpful as Mistika goes onto a broader range of systems. It's one corner of the UI that you definitely have to interact with frequently and that's cumbersome at the moment. I will keep adding to the shortcut spreadsheet over time. Or maybe if that is hard to collaborate on, we can start a forum thread where people can report helpful shortcuts they've found (or created) and then we summarize in a few weeks/months.

Obviously SGO should weigh in on this. But I'll give you my perspective, having first started on MacOS and now having moved to Windows recently. First of all, I don't think Mistika is Windows centric. They come from a world of Linux, not Windows. What they do focus on is NVidia cards, and since those are not available on Macs for the most part, it creates the perception of a Windows bias. In fact I would say Mistika has a bit of a rough time on Mac to live within the Mac desktop. All the system settings you have to adjust to have two screens, hide the menu bar. It's a bit painful. Mistika supports ProRes renders on Windows, so there is no issues there. Right now I'm running with two RTX 2080Ti cards and except for 8K footage my UI has been pretty much real time and easy to work with. Last night I was rendering out some dailies (HD, Varicam Log, basic LUT & grade) and my render performance was about 195fps. It took less than an hour to render out 320GB of dailies into ProRes HQ files on Windows. The new MacPro looks very nice. But if it's a system that you primarily use for Mistika, I would definitely go the Windows route because it gives you more GPU selection and the interface will be a bit easier.

A few days ago I spent some time in the hotkey editor of Mistika (accessible via the login screen) and discovered a myriad of interesting keyboard shortcuts, some of them quite useful. I started compiling some of the key ones for my own reference in a Google Sheet. Just a fraction of it, but gives you a flavor. Today I was looking to see if there's a better way to scroll up/down in the parameter tab on the left in the Visual Editor. I know you can scroll with 'right click - scroll'. That works great with a mouse or the pen. But doesn't work on the Wacom in touch mode so well. So I was looking for any keyboard short cuts to scroll the parameters up/down, and in the process discovered something even better: Ctrl+PgUp toggles full-screen on the on the bottom half of the visual editor. It hides the image and clips and expands the graph editor and parameter tab to full-screen (see screenshot). That's fantastic. So you can see everything at once while looking at your reference monitor .... Now that I know this feature, I realize there is a 'Max' button in the top right of the pane I never noticed. It does the same. But keyboard is faster. We should collect more of these useful keyboard shortcuts. But if someone know the one to scroll this up/down via keyboard, that would be great.

Now that I had the right keywords, I think I have finally found what I was looking for. See: http://developer.download.nvidia.com/GTC/PDF/GTC2012/PresentationPDF/SS101A-Scaling-GL-Transfer-Rendering.pdf On slide 10 it refers to GeForce and low-end Quadros having only 1 CE (Copy Engine) while Quadro 4000+ has two CEs.

OK, will do.

Thanks Javier. Right, that's exactly what the SGO article also states. What I was looking for is some documentation on that. I haven't been able to find anything from NVidia or elsewhere on the web. I'm using this article to justify a much more expensive card. And a lot of folks are skeptical since so many other comparisons call the cards equal. So I'm looking for some more information to back this up. I should have my card for evaluation soon so I can do some first hand testing of both of them in parallel. But if this is a key feature, I would expect NVidia to document that somewhere.

Not that I know off. But I haven't checked the tutorial videos on that lately. There is at least one of them that talks about it. The work around would be to quickly add a history (mapped on the panel) to preserve the current. Then you can cycle through the different presets with 'replace' and if you don't like it just restore the history as an undo. As an aside, one handy feature is the 'recover snapshot' in presets. It takes the selected preset and loads it as your snapshot. That doesn't apply it to the live shot, but is very useful for a side-by-side comparison. Even more useful if you have gallery shots you want to use as reference. Simply copy your .jpgs on the timeline temporary, create presets from them, then delete them, and now you can use these presets as reference wipes.

I currently use two GeForce RTX 2080TI cards but will be testing a Quadro RTX 6000 as a possible upgrade. This is largely based on this SGO support article https://support.sgo.es/support/solutions/articles/1000247927-nvidia-quadro-or-geforce- is there any documentation that is available that goes deeper into the bus management of the two cards. In almost all online content the two cards are considered largely identical in architecture and differences are said to be minor. This article stands out it saying the Quadro cards are far superior. I would love to find out more. Is there NVidia developer manuals or anything like that? Couldn’t find anything via Google. Of course I’ll see some performance data soon as I run a few projects against each other.

Or you can do both ? I used to be Windows for more than a decade, then 8 years of Mac only. And as of last month one of each.... I like the performance and flexibility of my new Windows system. I hate its usability. I keep using my Mac keyboard on my PC, and I use remote desktop to access various things on the Mac without having to switch screens and keyboards. Back to the original topic - we did find the root cause of the 'Save Image' not creating a file. Apparently the 'Save Image' only works if 'Live Video' is disabled. If you have Live Video enabled it will ask you for a filename but never actually create the file. Presumably it just doesn't have a rendered buffer at hand to actually save?

Thanks Cristobal. I'll try that out via the display filters.

I had problems with it creating DPX files. It doesn't work at all for me at the moment. That is on a Windows machine. I already have a ticket open for that and Laureano is working on it. We're trying to reproduce it. It shows me the file dialog but then no file is created.

Quick update: I've been able to work with a single project file and just switching mConfig for the various renders. The processing and rendering resolution seems to be tied to the config, so in a multi-resolution output setup varying the config works. That made copying grades between shots simpler. Of course having to exist Mistika, change the config, and restarting is still a slow process. Maybe in some future version we can make at least some config changes without restart....

Sure, I've started using Workflow for some things. But if I just need to grab a frame to send to the client and don't want to do a screen grab which is not as color accurate as Mistika saving a properly encoded file, having to fire up Workflow is quite the overhead.

I already make that feature request.