Nvidia is today dotting the i’s and crossing the t’s by confirming launch plans for a trio of imminent RTX 4060 Series graphics cards.
Following months of leaks and speculation, the Ada Lovelace architecture officially makes its mid-range debut on May 24 with the arrival of GeForce RTX 4060 Ti. Priced at $399 / $389 for a Founders Edition model, the mainstream card promises to deliver “unparalleled performance at fantastic value” and will be joined by two additional x60 cards in a couple of months’ time.
A 16GB variant of the regular 8GB RTX 4060 Ti will arrive in July priced from $499 and available exclusively from Nvidia partners. At the other end of the x60 scale, a base RTX 4060 will also launch in July priced at $299 and employing a purpose-built AD107 GPU.
The importance of this segment cannot be overstated. Steam’s most recent hardware statistics highlight Team Green’s dominance in the market, and crucially, four out of the five most popular GPUs are in the 60-class of previous generations. Nvidia is desperate for those folk to migrate to a newer 4nm model, and RTX 4060 parts are being introduced across three key price points in an effort to entice the many gamers still clinging to 10- or 20-series gear.
|GeForce RTX||4090||4080||4070 Ti||4070||4060 Ti||4060|
|Launch date||Oct 2022||Nov 2022||Jan 2023||Apr 2023||May 2023||Jul 2023|
|Die size (mm2)||608.5||378.6||294.5||294.5||187.8||TBC|
|SMs||128 of 144||76 of 80||60 of 60||46 of 60||34 of 36||24 of 24|
|Boost clock (MHz)||2,520||2,505||2,610||2,475||2,535||TBC|
|Peak FP32 TFLOPS||82.6||48.7||40.1||29.1||22.1||TBC|
|Memory size (GB)||24||16||12||12||8||8|
|Memory bus (bits)||384||256||192||192||128||128|
|Memory clock (Gbps)||21||22.4||21||21||18||17|
|L1 cache (MB)||16||9.5||7.5||5.8||4.4||3.1|
|L2 cache (MB)||72||64||48||36||32||24|
|Launch MSRP ($)||1,599||1,199||799||599||399||299|
Perusing the Club386 Table of DoomTM highlights sweeping changes in order to hit the target MSRPs. RTX 4060 Ti represents a 26 per cent reduction in core count compared to RTX 4070, and there are equally savage cuts at the back end, where memory bus shrinks to 128 bits and memory capacity is reduced to 8GB.
Available VRAM has become a hot topic in recent months, with certain high-profile titles demanding more memory at higher resolutions. Nvidia counters with 32MB of onboard L2 cache – a greater than 10x increase over RTX 2060 – which reduces the demand on memory subsystem to a certain extent.
Much to the dismay of add-in-board partners hoping to cash-in on demand for RTX 4060 Ti 16GB, Nvidia’s internal numbers hint at barely any discernible performance increase when playing a variety of games at a 1080p resolution with maximum quality settings.
We’ll await internal Club386 benchmarks to determine whether or not 8GB is sufficient, but it is important to reiterate target resolutions. Nvidia’s product stack is segmented in such a way that RTX 4080 and RTX 4090 are geared for 4K UHD, RTX 4070 Series is intended for 1440p QHD, while RTX 4060 Series aims squarely at high-framerate, 1080p FHD gaming.
On paper, RTX 4060 Ti appears an easy win for Nvidia. Pricing remains consistent with previous-generation RTX 3060 Ti, performance is enhanced through a latest-generation feature set comprising DLSS 3 and third-generation RT cores, while total graphics power is particularly frugal at just 160 watts.
RTX 4060 regular, meanwhile, is trimmed right the way back to a lightweight 3,072 cores. The 8GB card will go head-to-head with rival Radeon RX 7600 and if AD107 specifications look familiar, that’s because RTX 4060 has already made its debut in wattage-restricted laptop form.
We’re all set for a fascinating month of mainstream PC gaming. Watch this space, Club386 benchmarks and analysis coming soon.