Linus Sebastian: “I may simply skip RTX 4000 series”
Youtube / LinusTechTipsLinus Sebastian of Linus Tech Tips fame tweeted that he may “skip” the RTX 4000 series. While no reasons were directly given, we’ve got an inkling as to why.
The RTX 40-series is finally here, with the advent of the extremely powerful RTX 4090. In reviews, the card smashed through any and all performance scores of any graphics cards that we’ve seen before it. However, while the grass may indeed be green for Nvidia when it comes to pure performance, there’s more going on under the hood.
Linus Sebastian, recently expressed his own opinion, where he said that he may skip the entire generation. But, we’ve got an inkling as to some of the reasons why YouTube’s biggest tech nerd is pensive on the new GPUs.
The RTX 4090 is handicapped by its own display standards
The RTX 4090 is quick. It’s the fastest GPU you can buy by a mile. However, Nvidia curiously still uses the DisplayPort 1.4 format on the output of the 4090. At 4K, the RTX 4090 can achieve up to around 130FPS in titles like Cyberpunk 2077 at 4K, Ultra with all the Ray Tracing bells and whistles turned on , as we’ve tested on our review unit.
However, DisplayPort 1.4 is limited to 4K, 120Hz signals, meaning that all that power simply gets left on the table, since Nvidia did not choose to opt for a newer standard for the GPU, which is clearly capable of pushing more frames than that.
This is something that we’ll touch on in our upcoming RTX 4090 review, especially since there are rumors swirling that AMD may choose to opt for DisplayPort 2.1 for its RDNA 3 graphics cards.
The 4090’s power target is incredibly high
Another reason why Linus may choose to skip the current generation of GPUs is its power targets. The RTX 4000 series has some pretty hefty power requirements, especially at the higher end of the stack on the RTX 4090. However, it’s been observed by many, such as Der8auer. That this may simply be too high. You can easily reduce the power target by around 33% and sacrifice just 10% of the performance. This is illustrated through his in-depth testing.
The whole RTX 4080 debacle
The RTX 4080 is coming in two flavors, a 16GB model and a 12GB model. However, in reality, these are two wildly different graphics cards. Why Nvidia has chosen to do this and position these cards in such a manner remains to be a bit of a mystery.
In its own published benchmarking data, Nvidia’s RTX 4080 12GB is a paltry 12% faster when compared to the RTX 3080, with the 16GB model 30% faster than the 12GB. This is truly mind-boggling behavior. It doesn’t really sit well with us, and possibly Linus, too, that Nvidia is masking a 4070 as a “4080 12GB” model.
If this is true, what does the lower-end look like? Is this positioning supposed to entice customers to rush out and buy the 30-series? If so, the strategy may just be working.