NVIDIA are now transcending what it truly means to be a hardware company by being the very first of its kind to “unlaunch” a product.
No real surprises here, NVIDIA look to be washing its hands clean of its disastrous unveiling of the RTX 4080 “12GB” model or more accurately, what probably would’ve, should’ve, could’ve been the RTX 4070.
Obviously the backlash from the GeForce Goyim community was larger than imagined, for NVIDIA to backtrack on at least SOME of its blatant Jewishness the last minute name change obviously hasn’t gone over well with consumers or probably investors.
NVIDIA had announced its latest generation of GeForce GPUs, with the RTX 4090 officially released now it has shown to be roughly ~60% faster than an RTX 3090Ti on average, around the general line of where I gauged it to be from NVIDIA’s deceptive marketing.
However the real deception came in the form of the RTX 4080, where NVIDIA had announced two SKUs, one being 16GB, built on the AD103 core containing 9728 CUDA cores while the 12GB “variant” shares nothing in common with its stablemate as that is built on the AD104 core and contains 7680 CUDA cores in total.
“You will never be a real RTX 4080. You have no AD103 die, and you have 12GB of VRAM. You are an anemic 4070 twisted by marketing into a crude mockery of a flagship GPU.
All the “validation” you get is by soyjak tech YouTubers and prebuilt PC consumers. Behind your back people mock you. NVIDIA’s engineers are disgusted and ashamed of you, and your own AIBs laugh at your ghoulish performance behind closed doors.
Gamers are utterly repulsed by you. Decades of experience have allowed gamers to sniff out frauds with incredible efficiency. Even mislabeled hardware that “passes” looks uncanny and unnatural to a gamer. Your core count is a dead giveaway. And if you manage to run a game at 4K 120 fps, he’ll turn tail and rip you straight out of your PCIe slot the second he gets a glimpse of your glorified midrange performance.
Eventually it’ll be too much to bear – the marketing team will hold a meeting, gather some executives, and rename you as the 4070 Ti. The engineers who created you will be heartbroken but relieved that they no longer have to live with the unbearable shame and disappointment. They’ll ship you in iBuyPower PCs with the 4070 Ti badge, and every consumer for the rest of eternity will know a midrange GPU lies inside. When your cooling shroud is stripped away, all that will remain of your legacy is a die that is unmistakably that of an AD104.
This is your fate. This is what you were inflicted. There is no hope for you.”
Figuratively performance wise, the RTX 4090 is 60% faster than the previous generations 3090Ti, while the 16GB 4080 is presumed to be around 20% over the 3090Ti while the so called “RTX 4080 12GB” will undoubtedly be around 15% off the pace and presumably around the cusp if not slower than an RTX 3090.
Anyways, the real kicker of the RTX 4000 series came in the form of its price tag, a knock-on effect perpetuated by mindless consumers having purchased the GTX Titan graphics cards over the years, allowing NVIDIA to boil the frog slowly as they increase individual SKU price tags with subsequent generational releases.
And not just that but the company then went on to release new SKUs to bridge even more gaps, and to charge even more obscene amounts for products that otherwise should not exist, such as the X70-Ti and X90 / Ti class of graphics cards, with prices skyrocketing the dumb shits continue to buy these products causing NVIDIA to get even more greedy.
But now you can look forward to the barbaric $1600 RTX 4090 which actually costs upwards of $2500 in stores and the bullshit $1200 figure for the RTX 4080 16GB.
Sad to say we won’t get our wish of a $900 USD RTX 4070, because NVIDIA have pulled the plug on it
NVIDIA proclaims that the RTX 4080 12GB was “confusing” having two GPUs with the same designation is confusing, but I’ll tell you what wasn’t confusing, the RTX “Super” moniker.
If we turn the clock back a couple years further back to the GTX 1000 series you’ll notice that there’s several individual variants of the GTX 1060, specifically one with 6GB, another with 5GB exclusive to the Chinese markets, another with 3GB with even fewer CUDA cores, a GTX 1060 utilizing 9Gbps GDDR5 memory, a GTX 1060 utilizing defective GP104 dies and another one of those defective cutdowns with 6GB of GDDR5X.
NVIDIA doesn’t give two shits about being “confusing”, they are only taking action now simply because their Jewish tricks were met with universal criticism which for the green goblin is especially rare.
Receptively, ADA Lovelace has been a “success” with the RTX 4090, as consumer fangays ignore the obvious downsides, such as the sheer size of the 500+ watt behemoth, the bullshit fake MSRP costs and the new and innovative DLSS 3.0 which rapes your screen with interpolated frames.
Faggotry JewTube shills have been misrepresenting the FPS gains made with DLSS 3.0’s interpolated rubbish, as the frames themselves are generated as I previously mentioned on the announcement article this does in fact add lots of latency.
Hardware Unboxed’s video on the matter certainly has gotten the shills riled up for exposing the truth, gay bondage models such as Alexander Battaglia from Digital Foundry got BTFO’d.
As it’s an interpolation of what the next frame SHOULD BE, DLSS 3.0 does not generate in-engine rendered frames, NVIDIA of course market this cancer as being “similar latency to native rendering” just like how DLSS is “better than native” according to Digital Foundry.
Theoretically in some titles it could offer lower latency, if you were to magically double your framerate, but every other frame in the queue cannot display changes to input between the source and interpolated frames, it doesn’t matter of you’ve got double the FPS on screen, you’re effectively still running at half the framerate in terms of latency.
Discrepancies can occur when testing, such as when Digital Foundry jerks off over still images, because with DLSS 3.0’s frame interpolation the moment you click a button during a RENDERED FRAME you have to wait for two frames minimum for it to possibly display in game.
NVIDIA consumers are really the biggest sheeple on the entire planet, blatantly ignoring reality because Jensen Huang is selling them “magic”.
NVIDIA will more than likely backtrack in the sense that they will announce a proper RTX 4070 graphics card with fewer CUDA cores than the RTX “4080 12GB” and possibly offer anywhere from 8/10GB of VRAM as the RTX “4080 12GB” will now probably be reserved for an eventual 4070Ti.