Reshaping Workflows with Dell Pro Precision and NVIDIA RTX PRO GPUs
Sep 17, 2026 · 13 min · 10 segments
What’s slowing down your creative process—long render times, slow client revisions, or security concerns with cloud AI? In this episode of *Reshaping Workflows with Dell Pro Precision and NVIDIA RTX…
Jonathan WinbushGuestLogan LawlerHostYou have the Dell Pro Max T2 tower, uh, and you've done some crazy stuff with it.
But- The VRAM in ADA from kind of gen over gen in the discrete, you know, RTX kind of cards, like in the, the 6000, right, in the desktop, has went from 48 to 96.
And kind of gen over gen, like, what does that 96 gigs kind of unlock for you? Like specifically like what you couldn't have done before.
So when I first initially got the tower, I thought this was gonna open up my Unreal Engine workflow exponentially, because a lot of times you do hit that VRAM issue if you're using a lot of stuff within your view port.
Like if you don't optimize your scene, a lot of times we're throwing stuff at it, and then you'll hit that VRAM issue where it starts slugging and it's not real time anymore.
But honestly, once I really started getting into Comfy UI and doing localized AI generation, that's when I really saw a big difference, especially with some of these bigger models.
Like I like to use LTX, and they just released 2.5 the other day, in which before that I was using 2.3, and I could say for some of my video generations they might take like maybe like two to three minutes.
But with the latest version, especially how it's optimized for the NVIDIA card, literally I ran some tests and those same videos that I was running before is happening in like 33 seconds, which is insane.
And before, like you would hear the video card spinning up and everything, but now it's just like, it's dead silent.
I would say like depending on what I, the video that I'm trying to generate, I mean sometimes I've seen upwards up to five minutes, but typically it would be like between two to three minutes, and now it's like 33 seconds.
You have the Dell Pro Max T2 tower, uh, and you've done some crazy stuff with it.
But- The VRAM in ADA from kind of gen over gen in the discrete, you know, RTX kind of cards, like in the, the 6000, right, in the desktop, has went from 48 to 96.
And kind of gen over gen, like, what does that 96 gigs kind of unlock for you? Like specifically like what you couldn't have done before.
So when I first initially got the tower, I thought this was gonna open up my Unreal Engine workflow exponentially, because a lot of times you do hit that VRAM issue if you're using a lot of stuff within your view port.
Like if you don't optimize your scene, a lot of times we're throwing stuff at it, and then you'll hit that VRAM issue where it starts slugging and it's not real time anymore.
But honestly, once I really started getting into Comfy UI and doing localized AI generation, that's when I really saw a big difference, especially with some of these bigger models.
Like I like to use LTX, and they just released 2.5 the other day, in which before that I was using 2.3, and I could say for some of my video generations they might take like maybe like two to three minutes.
But with the latest version, especially how it's optimized for the NVIDIA card, literally I ran some tests and those same videos that I was running before is happening in like 33 seconds, which is insane.
And before, like you would hear the video card spinning up and everything, but now it's just like, it's dead silent.
I would say like depending on what I, the video that I'm trying to generate, I mean sometimes I've seen upwards up to five minutes, but typically it would be like between two to three minutes, and now it's like 33 seconds.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.