Let's be real, the only reason the 3090 isn't a Titan card is because Nvidia saw this coming and didn't want AMD to be able to top their "untouchable" Titan brand.

>3090 is actually a Titan without the Titan name.

AdoredTV made an entire video[0] to demonstrate this is absolutely baseless conjecture that NVIDIA Marketing tricked people into inferring.

It absolutely destroys the Titan idea, leaving no room for doubt.

[0]: https://www.youtube.com/watch?v=s23GvbQfyLA

I've gotta watch 30 minutes from some tech tuber to get your point? C'mon dude.

I was curious so I watched. It's a good video. I ended up going with the 3090 because of the extra vram and I'm familiar with CUDA but I'm not under the impression that this thing is a "Titan" even if it is all arbitrary corporate market segmentation. Relevant bit is at 15 minutes: https://www.youtube.com/watch?v=s23GvbQfyLA&15m1s

People who hate video can process a 4 page article in 15 seconds rather than minutes

There is no reason AMD wouldn't pull their own "Titan"-class GPU. A 6900XT with 32GB of HBM would destroy anything NVidia could muster.

I mean they kinda already did put out a Titan class card. My point was that the 3090 is a Titan in all but name, and AMD unveiled their very capable response today.

I'm saying the only reason Nvidia didn't call it a Titan was because they knew it wasn't going to have an unassailable advantage over AMD.

There's no Titan-"class"

A Nvidia card is Titan if it has the name and the drivers, and not if it doesn't.

You're no fun.

AMD Titan-class cards tend to have HBM. I'm saying that AMD can very probably still do even better.

AMD has a CDNA in their pipeline which they haven't talked about in a while. That's their compute oriented architecture, whereas RDNA is graphics-oriented.

Indeed. I still think it might be possible to see an RDNA2 HBM card.

I thought Quadros were meant for non-gaming workloads and the Titans were just more powerful than the top-tier mainstream/gaming card? Or have things changed?

Yes, Quadro was for non-gaming workloads. Titan meant a different thing with every new iteration. Some were explicitly "non-gaming card that you can play games on" other were "got tier gaming card".

RTX Titan for example was not as good in FP64 calculation compared to RTX Quadro. While first Titan was good at both. Nvidia kept changing what Titan means in order to extract most amount of money without hurting Quadro sales.

Even at its peak certain things were software limited for Titan, but not for Quadro. Nvidia has tons of artificial limitations for each of their graphics card once you go beyond gaming.

There's no titan this time around and nvidia have been talking about simplifying their lineup and implying the 3090 is to fill the titan shaped hole in the lineup, but also leaving them room to release a RTX Titan Ti or whatever if they change their mind. Also I imagine to head off consumers who would otherwise be like "Maybe I'll wait for a 3080 Ti".

I also expect been a kind of struggle between their consumer side who need some other reason to sell their top end GPUs beyond just the whales when game developers aren't really that interested in pushing out features that will require it given the niche ownership, while the datacenter people want to protect their Quadro margins. This to me seems their most likely reason for the on-again/off-again relationship with pushing their high end cards for compute usage vs gaming usage.