I took one of the two TitanX GPUs out and the steppyness seemed less but still unusable. If I run Resolve 15.3.1 there is no steppyness with either panel and they work on both the 10.12.6 and the 10.13.6. So Resolve 16.0 and up to 16.2.6 have the steppyness.
I'm aware that nVidia support on Mac OSX starting at 10.13.6 is flawed. Even if I manually set Resolve to only use the GTX-980 Ti for processing if the expansion chassis is on the steppyness occurs in Resolve 16.X, I tried both 16.0 and 16.2.6 and one in the middle of the pack. So it would seem the most viable option is to stick with my 10.12.6 startup which will only allow me to install Resolve 15.3.1 or other versions of 15.X because it will also launch Avid with the nVidia webdrivers on. I've read that the GTX-980 Ti 6GB is basically a TitanX with half the Vram but I am wondering if I had matching GTX-980 Ti GPUs in the expansion chassis if that might make things play nicer. I learned from Dom that Resolve will take the GPU with the least amount of Vram and I apply that limit to all other GPUs. That seemed to ring true in my tests. So I wonder if it's the mere presence of GPUs in the expansion chassis that causes the problem or if the absolute type of GPU. It was pointed out in this thread that my original GTX-680 was a Kepler chipset and the TitanXs are Maxwell chipsets. That's part of the reason I went with the GTX-980 Ti because it's a Maxwell chipset. Clearly there is more going on. I might try putting the GTX-680 in the expansion chassis to see if the steppyness re-occurs. Of course that wouldn't make any practical sense because then Resolve would only use 2GB Vram on both GPUs for a total of 4GB but it might point to the possibility that perhaps GTX-980 Tis in the expansion chassis would work better.
The only other factor I can think of is the expansion chassis is connected on a 3M pigtail to the tower. I also have a 1M pigtail. The manufacturer told me both lengths meet spec and there should be no performance issues with the longer pigtail. I'm just thinking perhaps the cable delay due to extra length might also be a factor. It would be physically difficult to use the shorter cable but if I put on my big boy pants it might be worth a shot.
It seems the God's of OS/Technology change are not in my court on this stuff.
On Sat, Sep 12, 2020 at 09:08 AM, editbruboy wrote:
I have a setup with Resolve 15 on Win 7 and the Artist Color is working fine with Eucon 3.6.1
Reading your plight made me decide to see how well the Artist Color would work across the network on another machine.
I therefore installed Eucon 3.6.1 on a laptop with Win10 and Resolve 16.2.1.
Apart from an alert message from my antivirus software about one of the Eucon components (which I allowed) the install worked fine and I could see/assign my panel in the Eucon software.
I'm pleased to report that the Artist Color seems to behave exactly the same. Haven't done extensive testing, but using the three trackballs and rings seems about as responsive as I remember.
Obviously, this is on a Windows setup so I don't know what's happening with your Mac hardware/software. The only thing that is different is that I'm using an earlier version of Eucon. I don't think the later versions added any functionality to the Artist Color and was mainly aimed at improving their audio surfaces. I did read that someone had problems with their Artist Color when they (naturally) upgraded their Eucon software and went back to the earlier version.
3.6.1 might be worth a try.
Bruno
On 12/09/2020 01:30, John Moore via groups.io wrote:I've been beating my head trying to test my Artist Color Panel in Resolve 16 OS 10.13.6. It's just awful and I don't remember it being that way before in my classes etc... Today I went back to my 10.12.6 startup Resolve 15 and the panel is very smooth like I remembered. Has anybody noticed the degradation of the sensitivity/perrformance of the Artist Color Panel on Resolve 16. I'm using the same Eucon software on 10.12.6 and 10.13.6 as that hasn't been updated because the panel is EOL. I can't install Resolve 16 on OS 10.12.6 so I can't tell if it's Resolve 16 or Eucon/10.13.6 issue. I wish Resolve would at least let me risk installing Resolve 16 on 10.12.6. I know this is old hardware but it works fine on the 10.12.6 startup.I'm going to borrow another Artist Color Panel from a friend to doube check and also hist BM Mini panel. It really feels like Resolve 16 may have crippled the Artist Color even though it is still an option in the pulldown menu. Or it's OS 10.13.6/Eucon? Nothing like herding software cats.Anybody seeing what I'm seeing?John Moore Barking Trout Productions Studio City, CA bigfish@pacbell.net
Groups.io Links:
You receive all messages sent to this group.
View/Reply Online (#134813) | Reply To Group | Reply To Sender | Mute This Topic | New Topic
Your Subscription | Contact Group Owner | Unsubscribe [administrator242.death@blogger.com]
No comments:
Post a Comment