Another update. (This thread is morphing into griping about Tech Support.)

I got a 2nd response from Tech Support. This time they wanted me to send proof of purchase.

Why? What difference could this possibly make? I made a video of PD 18 running on my computer. I showed the version number. So we know it really is PD18.

I showed a SN in the video I made for CL. I gave CL my Activation Key when I submitted my Ticket. Suppose I had driven to JL_JL's house and made a video of this problem occurring on his licensed version of PD18? It doesn't change the fact that the problem exists, no matter who paid for the license.

I don't believe that Tech Support has sent my video to Engineering.

This is getting silly.

(Moderator: I don't know what kind of computer you have. But if you have one with an Intel iGPU and an Nvidia GTX, perhaps you can replicate my results and make your own report to CL? They might listen to you.)

Quote I showed a SN in the video I made for CL.

FYI what you pixelated in your video is the Software Release (SR) number, which isn't a serial number of any kind. Also, remember that there's likely a language difference between you and the people in tech support, and what I've found is that keeping the focus on one or two points at a time works better than trying to explain everything in detail. YMMV

After watching your video it's clear that issue can be easily resolved with HDMI switch boxes as I mentioned in my earlier thread. In my experience, PD's GPU selection doesn't necessarily match what you've told Windows to do and it can easily give conflicting options when you're running one monitor off of each GPU.

Without a switchbox, if you were to physically connect both monitors to your GTX-1650 SUPER I'll guarantee you'll have full use of the NVENC.

Since I have 2 switchboxes, I can swap QuickSync for NVENC in seconds by switching both monitors at the same time with PD closed. If I leave one on the UHD 630 and one on my RTX 2070, I can actually get PD to show me the nVidia icon on the About screen but have the QuickSync option on the Produce page.

Long story short: it would certainly be nice to be able to control PD's GPU use via the OS tools, and it would be great if all your effort here resulted in a patch. Practically speaking, however, you can solve this right now with a minimal investment or even just a simple physical connection change.

Tnx. I thought it best to play it safe by pixelating what I thought was private info in the video. No harm, no foul.

As for language differences - yeah, I wonder if I have to break thru the "Saving Face" element I've read about in Oriental Cultures. I've run up against it with Germans, who don't believe that they ever do anything wrong.

Yes, you're correct, that I could work around the problem now with an HDMI switch. (Or the "Disable the iGPU in Device Manager" Trick.) It's not a big deal for me at this time for what I do with PD. And, considering the benchmark someone did here the other day, we now know with certainty that almost any change to a video proper forces CPU rendering instead of NVENC.

I know that some here believe that this is not a PD problem, per se. Whereas I believe it is. (For reasons I've stated before.) Given my current belief, coupled with my belief that it can be easily fixed in the UI section of their code (where I am guessing no one has ever looked for this problem), I'm trying to push it for now, for both the sake of users AND for Cyberlink. (I say this an an Engineer myself, who knows the (albeit) intuitive Eureka! feeling just before I finally fix a problem.)

But if my video doesn't get pushed up to Engineering soon, where someone can step thru the code and debug (and either fix it or tell me I'm wrong), I'll drop this. At least I'll know I tried.

Quote
Since I have 2 switchboxes, I can swap QuickSync for NVENC in seconds by switching both monitors at the same time with PD closed. If I leave one on the UHD 630 and one on my RTX 2070, I can actually get PD to show me the nVidia icon on the About screen but have the QuickSync option on the Produce page.

Yes, exactly what I showed in my video. But as I also showed, when I forced Windows to show PD 18 only the Nvidia GPU, the QuickSync option on the Produce page was still showing, even though it was not a valid (that is, not selectable) option anymore.

Quote Yes, exactly what I showed in my video. But as I also showed, when I forced Windows to show PD 18 only the Nvidia GPU, the QuickSync option on the Produce page was still showing, even though it was not a valid (that is, not selectable) option anymore.

That's almost certainly because of how Windows and/or PD dealt with the disabled iGPU, essentially leaving some amount of its presence cached.

If you rebooted and started PD again, the issue wouldn't be present. That might also be true if you were to close PD before making the monitor changes you made in your video.

Quote [bcolor=#ff9900]
That's almost certainly because of how Windows and/or PD dealt with the disabled iGPU, essentially leaving some amount of its presence cached.

If you rebooted and started PD again, the issue wouldn't be present. That might also be true if you were to close PD before making the monitor changes you made in your video.

That's an interesting theory. Is that what happens when you use your HDMI switcher?

If it were a cache problem, then it seems like a bug again, with the solution to clear the cache upon restart of PD. (I notice now with PD 18, there is a "pause" after it starts, where it seems like it is thinking about something. Perhaps GPU hardware? That's a change from PD15 to 18 for me. Perhaps 16 and 17 have a similar pause?)

I will try a reboot of Windows after the change to confirm whether it works or not. (While I don't recall rebooting Windows for this specific test, I know that I rebooted Windows several times while making my video (to change BIOS, etc.) and nothing changed.

But I'll try it later tonight when I'm done using the computer for other tasks.

Quote That's an interesting theory. Is that what happens when you use your HDMI switcher?.

As long as PD is closed when I switch monitors, I don't have to reboot. It takes Win10 a few seconds to adapt to the new configuration and then it's clear to launch PD.

I made a 3 minute video trying the experiment, as I understood it. Sadly, it did not work for me.

https://jmp.sh/VjylEst

Thanks for making the video, which told me in no uncertain terms that I need to work on being more clear when I type my responses!

I actually hadn't said anything about the Windows GPU control panel at all, and you can see I haven't made any changes to it:



When I mentioned possible monitor caching and guaranteeing that a reboot would give you the expected hardware option, I was talking about switching monitors using my external HDML switchboxes. In your case, that would mean physically connecting both monitors to your nVidia card, which should also override the Graphics settings controls.

Again, that's different from having PD and Windows communicate better but it is a way for you to reliably get full access to your nVidia GPU.

Ah! Well, in your case, the Windows' Graphics Settings defaults to "System." Which, of course, means than when you have nothing but your Nvidia driving your two monitors, PD has no choice but to "select" the Nvidia.

The reason that I have my Graphics cards/monitors AND Graphics Settings set up the way I do is because I also use Magix's Movie Edit Pro. And Magix made the (bad) decision years ago to optimize their editor for QuickSync. So when I run Magix, I have for be sure that Windows is pointing it to my UHD as the main GPU. Once that's done, Magix uses QS for Hardware encoding, but can also see my Nvidia for HA rendering during playback and even for some effects during encoding. So I need both GPU's operating.

I don't have Resolve on my box right now, but I believe that Resove is similar (but opposite?), in that it requires users to free up their Nvidia card for Resolve to use for encoding. And so one must be judicious with Graphics Settings with Resolve too.

(As an aside, when I finished my instant video, I began to wonder if you meant that PD caches the icons on the About Page, where it shows what video capabilities a user has. If yes, then I understand, and I, too, expect that those icons are cached, and won't change when video cards are changed until at least a restart of PD.)

Quote Ah! Well, in your case, the Windows' Graphics Settings defaults to "System." Which, of course, means than when you have nothing but your Nvidia driving your two monitors, PD has no choice but to "select" the Nvidia.

No, actually that's the case no matter how my monitors are connected: 2x iGPU; 1x iGPU + 1x nVidia; or 2x nVidia. All show the same control panel settings because I've never changed anything there.

I know you started out following the advice on other threads about laptops being unable to utilize the discrete nVidia card, and I guess what I'm saying here is that setting is irrelevant on desktops where you can access the individual GPU ports.

Regarding your overall workflow (and just out of curiousity), are you seeing this same kind of situation if you do have both monitors running of your nVidia card - Magix can't use the iGPU? If that's the case, then an HDMI splitter/combiner switch only costs as little as $8 on Amazon and it'll solve all your problems!

Quote
I know you started out following the advice on other threads about laptops being unable to utilize the discrete nVidia card, and I guess what I'm saying here is that setting is irrelevant on desktops where you can access the individual GPU ports.

Well, I know that you're the expert. But I don't believe that the Graphics Setting setting is irrelevant, because, as I show in my first video, when I tell Windows to tell PD 18 to use my Nvidia GPU, then PD 18 won't let me click on/enable/select the QuickSync radio button anymore. (And I argue that the QuickSync button shouldn't be there as an option to click after WIndows points PD to my Nvidia.)

So I infer that there is some GPU switching going on at the Windows level. I maintain that PD 18 isn't acting correctly on the switch. (I.e., a bug.)

As to your question about whether Magix can access my Intel iGPU if no monitor is hooked up to it. A: I haven't tried it. My BIOS has text that says the iGPU can be available to Windows when an external GPU is used. (Might be a Dell thing.) The Magix guys, in their forum, say one needs a headless adapter/ghost to make Windows think the iGPU is alive.

As an aside, I did look at HDMI switches on Amazon. I saw one for $10. I would need to buy another HDMI cable. (I have a DVI-D adapter, since my GTX only has one HDMI output.) So $20 for a workaround that I don't need at the moment. But it's on my background list of things to buy someday.

But I'm hoping that CL will fix this before then.

Quote But I'm hoping that CL will fix this before then.

No reason to go around in circles here, but I don't think that's very likely. As far as I'm aware, this issue has been present ever since I got a second monitor back in 2013 and was eagerly awaiting the new bells and whistles that PD12 would bring.

I'm going to start a new thread and document the monitor misidentification issues that I've seen for the express purpose of reporting the issue to CL, but my honest advice is to buy the switch and move on

Quote
I'm going to start a new thread and document the monitor misidentification issues that I've seen for the express purpose of reporting the issue to CL, but my honest advice is to buy the switch and move on

Glad to hear it. They have listened to you in the past. Perhaps they will listen to you again and you will find success.

Meanwhile, I'm still stuck in Customer Support getting the script runaround. The latest is that they wanted me to 1) update my driver and 2) send a screen shot of the Profile I'm using.

As for 1) I was using the latest driver when I sent in the report. Then a new Studio driver came out. I updated. But this is not a driver problem. So no change.

As for 2) I fail to understand how Profile can matter, since, for the Profile I am using, the Profile allows Hardware Acceleration for QuickSync, and then when I disconnect the iGPU, the same Profile allows Hardware Acceleration using the Nvidia. So in my instant case, this is not a Profile problem.

Argh!

I know how frutrating this is, and I've found it's better to give them the requested info even when it seems unnecessary. You and I don't know what they're looking at on their end, after all, and things that seem silly or irrelevant from the outside have occasionally turned out to actually matter.

Some encouraging news: I got a reply from Tech Support. They finally tried it and have reproducted this issue. My Ticket has been escalated to Engineering. (Ironically, they suggested, as a temporary workaround, optodata's finding: Run both monitors from the Nivida.)

I thanked Tech Support and referred them to Optodata's newest finding, giving them his Ticket # and the link to his discovery.