This collaborative space allows users to contribute additional information, tips, and insights to enhance the original deal post. Feel free to share your knowledge and help fellow shoppers make informed decisions.
Yes. That's the core purpose of the card. 32gb of vram. It's not fast, but it is one of the cheapest if not the cheapest ways to get a modern 32gig card at the moment.
If you've got a threadripper and a bunch of ram and a bunch of free PCIe slots and your system is going to be engineered from scratch and you're never going beyond your scope and you don't need anything NVIDIA specific? OMG! This is awesome! Though two 24GB cards cost about the same price.
These can actually game between a 9070 and 9070xt just with 2x the vram, if anyone is thinking about a dual purpose machine. But keep in mind only the Asus and Gigabyte ones have HDMI ports, and I have yet to find an active or passive dp to hdmi adapter that passes 4k/120 + VRR.
I went from 3090s to V100s to a strix halo 128gb, and have thought about swapping to 2x of these or just adding one to the chain (and hooking it up to the living room tv while it's idle)
If you've got a threadripper and a bunch of ram and a bunch of free PCIe slots and your system is going to be engineered from scratch and you're never going beyond your scope and you don't need anything NVIDIA specific? OMG! This is awesome! Though two 24GB cards cost about the same price.
Likely 3090s, though they are selling for 800+ (usually 900+)The 32gb blackwell 4500 was 2300, now 3300+. Previous gen ada 5000 is the same price, so nvidia is a big jump up for 32gb. V100s are solid value but very old architecture and no support for key features like flash attention
And for anyone thinking all this LLM stuff is going over their heads, don't worry. It's still very much the wild west.
Plenty of opportunity to learn and everything is still very much in the infant stages.
It's only going to get better for hobbyists as less and less people support the closed models OpenAI and Anthropic are putting out.
If you're confused about what to buy, this should make it easy: Nvidia is currently dominating gen-AI. If your goal is visuals and industry proven configurations, stick with them.
Intel ARC and Radeon are still pretty experimental and require more patience to set up, but they are catching up when it comes to text gen (coding, writing, etc).
Like
Helpful
Funny
Not helpful
Join The Conversation
Share your experience with the Slickdeals community
Share information with the community. Please follow our Community Guidelines and be kind!
Join The Conversation
Share your experience with the Slickdeals community
Share information with the community. Please follow our Community Guidelines and be kind!
14 Comments
Sign up for a Slickdeals account to remove this ad.
Sign up for a Slickdeals account to remove this ad.
https://youtu.be/efQPFhZmhAo?si=
I went from 3090s to V100s to a strix halo 128gb, and have thought about swapping to 2x of these or just adding one to the chain (and hooking it up to the living room tv while it's idle)
The 32gb blackwell 4500 was 2300, now 3300+. Previous gen ada 5000 is the same price, so nvidia is a big jump up for 32gb.
V100s are solid value but very old architecture and no support for key features like flash attention
https://youtu.be/4IotZmCQp-g
And for anyone thinking all this LLM stuff is going over their heads, don't worry. It's still very much the wild west.
Plenty of opportunity to learn and everything is still very much in the infant stages.
It's only going to get better for hobbyists as less and less people support the closed models OpenAI and Anthropic are putting out.
If you're confused about what to buy, this should make it easy: Nvidia is currently dominating gen-AI. If your goal is visuals and industry proven configurations, stick with them.
Intel ARC and Radeon are still pretty experimental and require more patience to set up, but they are catching up when it comes to text gen (coding, writing, etc).
Join The Conversation
Share your experience with the Slickdeals community
Share information with the community. Please follow our Community Guidelines and be kind!