Nvidia has announced the RTX 4090 and 4080 graphics cards, DLSS3

Nvidia has announced the RTX 4090 and 4080 graphics cards, DLSS3

At its GTC AI conference, Nvidia has announced its latest generation of GeForce RTX graphics cards. The RTX 4090 and RTX 4080 are the first GPUs from the ''Ada Lovelace'' series, and they should be up to twice as fast as their last-gen counterparts in rasterised games - and up to four times as fast.

The RTX 4090 24GB will be available on October 12th, while the RTX 4080 16GB will be available in November, and will cost $1199/1269. A 4090 and 4080 model is estimated to be two to four times faster than their Ampere counterparts, the 3090 Ti and the 3080 Ti.

RTX 4090 24GBRTX 4080 16GBRTX 4080 12GB
Die size611mm380mm380mm
CUDA cores1638497287680
Boost clock2.52GHz2.51GHz2.61GHz
Memory interface384-bit256-bit192-bit
Memory bandwidth1018GB/s742GB/s557GB/s
Power usage450W320W285W
PSU requirement850W750W700W
PSU cables3x 8-pin3x 8-pin2x 8-pin
Release dateOctober 12th, 2022November, 2022November, 2022

Note that the Nvidia RTX 30-series cards will remain on sale for the moment, filling in the bottom portion of the market until lower-tier RTX 40-series GPUs arrive.

So how are these 40-series GPUs so fast? Well, the TSMC''s ''4N'' process allows the generation to become significantly more power-efficient, although we expect the flagship cards to be as power-hungry as we heard - you''re only getting a ton of extra performance to mitigate that in terms of efficiency.

The new streaming multiprocessor uses a new technique, shader execution re-ordering (SER), which Nvidia claims to provide a "two to three times speed-up" for ray tracing, and a 25 percent increase in GPU resources. Similar advancements in the dedicated ray tracing silicon, with a doubling of ray-triangle throughput, a new opacity micromap engine that ''increases geometric abundance without the BVH build and storage cost.

Nvidia''s DLSS 3, a new technique that enables entirely new frames, is basically adding interpolated frames between''real'' ones to dramatically increase frame rates. This has a negative effect on input latency, and it is combined with Reflex to significantly reduce CPU and GPU charges, thus you will reap advantages in CPU and GPU-limited games, although the input latency penalty means it''ll be less effective for, say

Nvidia demonstrated Cyberpunk 2077 at a depth of 22fps at 4K with RT enabled and DLSS disabled, then 100fps with RT enabled and DLSS 3 engaged, a massive speedup even if this is a cherry-picked demo. This year, the game ran at 60fps with RT + DLSS 3, and 135fps with RT + DLSS 3 enabled; good evidence that the

Portal RTX is a new''remaster'' that Nvidia has previously called ''RTX Remaster,'' a software that allows you to develop mods for a wide spectrum of games that include ray tracing and AI-upscaled textures.

Nvidia has announced Racer RTX, a''simulated world'' with fantastic real-life - their marble demo is now up to 11.

Nvidia gave an excellent presentation, and we''re looking forward to testing both the performance and the features of the new generation. It''ll be fascinating to see whether DLSS 3 meets Team Green''s high promises... and whether these new cards will be offered at their previously announced prices if the performance concerns are raised. If the information is confirmed, we''ll leave more information here.

Related Articles