Page 90 - IBT October 2020

Tech

The Nvidia Nvidia’s new Ampere out of luck. announced that they would
reveals its
secret
Just as impressive as Nvidia’s
architecture, powering the new
acquire Arm, one of the world’s
arsenal, and
leading microprocessor
RTX 3000 series of GPUs,
hardware is their software,
companies, for USD 40 billion,
which utilizes an image
Rising gears up for that the RTX 2000 series cards processing technique by the it was immediately obvious
seeks to correct the problems
the age of AI
name of DLSS, or deep learning
posed. Manufactured using a
that this would become the
special 8nm process, the 3000
super sampling. This system
beginning of Nvidia’s foray
runs on the AI-driven Tensor
into entering the
series represents a phenomenal
Tide of compared to the previous used to clean up low-quality serious contender, because
leap in computing power
microprocessor market as a
cores of the RTX cards, and are
Arm-based CPUs lie at the
high-noise frame renders, and
generation of RTX cards. Even
upscale them to native
the ‘entry-level’ RTX 3070, the
heart of the vast majority of
smartphones, tablets,
featherweight of the 3000
resolutions, producing a final
Green superior to that of the USD that of a natively rendered certain laptops. Even Nvidia’s
integrated solutions and even
lineup, offers processing power
image of a quality similar to
own Tegra series of mobile
1400 RTX 2080 Ti, the
image. While this may sound
champion of the 2000 lineup –
like ‘cheating’ to most people,
processing solutions (notable
the end results have been so
at a flabbergastingly affordable
for powering the best-selling
starting price of USD 499.
quality, speed and efficiency,
gaming consoles) have been
While this is essentially extraordinary in terms of Nintendo Switch portable
By Abhijit Asad cannibalizing Nvidia’s existing that they have opened up based on processor
stock of RTX 2000 GPUs, the entirely new avenues for architectures licensed from
other cards on the current high-resolution raytraced Arm.
real-time renders By acquiring Arm, Nvidia
in games and would be able to make new
other 3D leaps and bounds in terms of
applications technological developments to
alike. The DLSS its own products, as well as
algorithms have the products of all its licensees,
been upgraded to the gamut of which
version 2.0 for encompasses every
the RTX 3000 smartphone manufacturer on
series, which has the planet. Especially with
resulted in a even major names like Apple
tremendous rise switching to Arm-based CPUs
Nvidia went for – by unveiling in the quality of for their future computers,
their new RTX 3000 series of the upscaled Nvidia could not have made
real-time raytracing-capable
Marvel’s Hulk is not the only GPUS. consumer-grade GPUs frames, to the point that some this acquisition at a better
raging green juggernaut In lay terms, raytracing is a equipped with real-time 3000 lineup, the RTX 3080 and of said upscaled frames look time. As their first move after
smashing things and turning method of accurately raytracing capabilities, were RTX 3090, are even better when compared to their the Arm acquisition, Nvidia is
scenes upside-down. The Big simulating the continuous launched with a great deal of heavier-hitting, and given that natively rendered counterparts. building a new AI research
Green, superconductor titan reflection and eventual fanfare back in 2018, to Nvidia has wisely opted to not Aside from graphics, the center at the UK-based
Nvidia, has not only revealed dissipation of light rays in a relatively mediocre sales – go with the eye-watering prices generational improvements to headquarters of Arm, with a
its latest range of absurdly 3D scene, leading to incredibly partially owing to the market’s of the RTX 2000 series, are the Ampere series of cards specific focus on developing
overpowered discrete graphics realistic lighting and general unawareness about the going to be a great deal more make them suitable for use in energy-efficient Arm-powered
processing solutions, but also shadowing. This is an importance raytracing can popular with PC enthusiasts deep learning-based solutions for data centers and
taken the world by storm by extremely resource-intensive make, but more importantly who want their money’s worth applications. Offering a 70% cloud services.
means of an acquisition process, with traditional because of the prohibitively when building their new speed boost over the previous The future has never looked
decision that has rocked the CPU-driven raytracing high price tags of the RTX systems. However, it should be series (Turing), Ampere can be brighter for Nvidia than it does
world of technology with its mechanisms requiring hours to 2000 GPUs. Only the most noted that because of their up to six times faster in AI now, and while it remains to be
seismic impact. spit out a single rendered high-end cards of the RTX vastly boosted processing training applications. seen how AMD and Intel would
With longtime rivals Intel and frame. While GPU-powered 2000 lineup could steadily capabilities, the While the revelation of the RTX respond to the new challenges
AMD working on their own renderers cut down on this handle real-time raytracing at Ampere-powered GPUs are also 3000 series cemented Nvidia’s posed by the green team, it can
processing solutions both for render time significantly, it over 60 frames per second, but quite power-hungry. The RTX position as the dominant be safely said that such
general and graphical was still a far cry from with their prices safely out of 3080’s power draw peaks out leader in the GPU market, with developments would have
purposes, it was apparent that achieving raytracing in real the reach of most buyers, they at a mammoth 320 watts, the competition yet to catch up massive impacts on the global
a do-or-die situation was likely time, at playable frame rates. did not see the level of which means users who do not to their benchmarks, this was computing sector as a whole,
to appear on Nvidia’s horizon Nvidia’s RTX 2000 series of adoption that Nvidia had have or are not ready to go for but one half of their and that the consumers would
before long. And ‘do’ is what GPUs, the world’s first initially hoped for. beefy power supply units are masterplan. When Nvidia be their biggest beneficiaries.
84