GTX 1660 Ti vs RTX 3060
The complete head-to-head: full specs, real per-game FPS with 1% lows, ray tracing, a live CPU-bottleneck check, thermals, power and value, with a straight verdict.
GeForce GTX 1660 Ti
GeForce RTX 3060
avg, 1080p
The short answer
This is a real tier gap, not a close race, and it's worth saying plainly: the RTX 3060 is about 57% faster on average, has double the VRAM, and adds ray tracing and DLSS that the GTX 1660 Ti simply cannot do. The one place the 1660 Ti holds its own is efficiency, its frames-per-watt is close to the 3060's despite being two years older. If price is the only constraint, the 1660 Ti is still a usable 1080p card. For anyone who can stretch the budget, the 3060 isn't a sidegrade here, it's a genuine generation and a half ahead.
These two get compared mostly by people on a strict budget trying to figure out how much more performance a bit more money actually buys. The GTX 1660 Ti was NVIDIA's last major GTX-branded card, Turing architecture without the RT and Tensor cores that made the RTX 20-series special. The RTX 3060 is a full generation newer, with real ray tracing hardware, DLSS, and, notably, the exact same 192-bit memory bus width paired with double the VRAM.
Since both cards actually share a 192-bit memory bus despite being a generation apart, the breakdown below leans on that: 1080p and 1440p frame rates, ray tracing and DLSS results, a CPU pairing tool, thermals, power, and price trends since 2023.
At a glance
Seven dimensions, two cards. Bigger area is not automatically better, each card owns different corners.
Strengths radar
Each axis scored 0-100 relative to the pair
Performance vs price
Up and to the left is better value
Grey dots are neighbouring cards for context.
Which should you buy?
Pick your use case, the winner changes with what you actually do.
Still capable of 1080p high in plenty of games, and the cheapest way into a working NVIDIA card on this page.
12 GB and meaningfully more raw compute make the 3060 the only realistic 1440p option of the two.
No Tensor cores at all on the 1660 Ti means the 3060 isn't just faster here, it's the only one that's actually usable.
The 1660 Ti has zero RT hardware. This isn't a performance gap, it's a missing feature entirely.
120 W and a 450 W PSU recommendation versus 170 W and 550 W, the 1660 Ti is the easier fit for a cramped or older build.
Both share a 192-bit bus, but the 3060's higher bandwidth and double the VRAM still give it the edge in demanding cores.
Own a 1660 Ti? This is one of the upgrades actually worth making.
Short answer: probably not
Unlike most sidegrade comparisons on this site, this one is a genuine upgrade. A 57% average gain plus real ray tracing, DLSS, and double the VRAM is worth $180 if you're still gaming on a 1660 Ti and want more headroom, especially for 1440p or anything AI-adjacent.
If you'd rather stretch further, an RTX 4060 or RX 7600 costs a bit more and adds AV1 encoding and a more efficient architecture on top of similar or better raw performance. But the 3060's 12 GB buffer is still the better pick if VRAM capacity matters most to you.
Full specification comparison
Every spec side by side. The highlighted figure wins each row.
| Specification | GTX 1660 Ti | RTX 3060 |
|---|---|---|
| Architecture | ||
| GPU / architecture | TU116 · Turing | GA106 · Ampere |
| Process node | TSMC 12 nm | Samsung 8 nm ▲ |
| Release date | Feb 22, 2019 | Feb 25, 2021 |
| Launch MSRP | $279 ▲ | $329 |
| Compute | ||
| CUDA cores | 1,536 | 3,584 ▲ |
| RT cores | None (GTX, no RT hardware) | 28 · 2nd gen ▲ |
| Tensor cores | None (GTX, no Tensor hardware) | 112 · 3rd gen ▲ |
| Base / boost clock | 1500 / 1770 MHz | 1320 / 1777 MHz |
| L2 cache | 1.5 MB | 3 MB ▲ |
| Memory | ||
| VRAM | 6 GB GDDR6 | 12 GB GDDR6 ▲ |
| Memory bus | 192-bit | 192-bit |
| Bandwidth | 288 GB/s | 360 GB/s ▲ |
| Power & form factor | ||
| TDP | 120 W ▲ | 170 W |
| Power connector | 1× 8-pin | 1× 8-pin |
| Suggested PSU | 450 W ▲ | 550 W |
| PCIe interface | 3.0 ×16 | 4.0 ×16 ▲ |
| Features | ||
| Upscaling | FSR 2/3, XeSS (no DLSS) | DLSS 2 / Super Resolution ▲ |
| AV1 encode (NVENC) | No | No |
| Display outputs | 1×DP1.4 · HDMI 2.0b · 1×DVI | 3×DP1.4a · HDMI 2.1 ▲ |
Average FPS by game
Switch resolution, metric, ray tracing and upscaling, the bars update live. Hover a bar for the avg / 1% / 0.1% breakdown.
Average FPS · 1080p · Rasterization. Representative data, hover a bar for avg / 1% / 0.1%. Note the RTX 3060's 1% lows fall off at 1440p and up from its 8 GB VRAM.
Smoothness & frametimes
Average FPS hides stutter. This is one run plotted as frametime, a flat line is smooth, spikes are hitches. Watch what the RTX 3060's 8 GB VRAM does in a texture-heavy game.
Frametime: Hogwarts Legacy, 1440p Ultra
Lower and flatter is better · 16.7 ms = 60 fps
Counterintuitive but real: the slower GTX 1660 Ti delivers a smoother line here because 12 GB holds the texture pool the 8 GB RTX 3060 keeps evicting, those spikes are VRAM stutter, not raw speed.
Lab scores
Clean apples-to-apples tests, useful, but weaker at predicting real games than the FPS above.
Ray tracing & upscaling
Ray tracing tanks both cards at native resolution. What makes it playable is upscaling, and here the RTX 3060 has a real edge.
Is frame generation "real" performance?
Neither card supports DLSS 3 Frame Generation, that needs Ada Lovelace hardware neither of these has. The RTX 3060 does support DLSS 2 upscaling, which helps frame rate without any interpolation risk. The GTX 1660 Ti has no NVIDIA upscaling at all, since it lacks Tensor cores entirely, it can only fall back to FSR or XeSS.
AMD's FSR 3 Frame Generation can technically run on both, but the 1660 Ti's low base frame rate in demanding games leaves little room for it to look good. Treat frame generation as a 3060-adjacent feature at best, and even there it's DLSS 2 upscaling, not true frame generation, doing the heavy lifting.
Will your CPU bottleneck these cards?
A faster GPU only helps if your CPU keeps up. Pick your CPU and resolution to see expected FPS and bottleneck for each card.
CPU + GPU pairing check
The willitbottleneck engine, built right into the comparison.
Estimated from a 43-CPU model across 1440p. A weak CPU mostly bites at 1080p; at higher resolutions the GPU becomes the limiter and the bottleneck shrinks.
Thermals & noise
The efficiency gap shows up here: the newer card runs cooler and quieter, day in and day out.
Real draw, efficiency & PSU
TDP is the rated figure; what matters is real gaming draw and how many frames you get per watt.
Both use a single 8-pin connector. If your PSU already ran the GTX 1660 Ti, it will run the RTX 3060 easily. Leave about 30% headroom for spikes.
Will it fit, and what to pair it with?
Size to scale
Card length matters for small cases
What monitor to pair
GTX 1660 Ti: best kept to 1080p at 75-100 Hz. RTX 3060: comfortable at 1080p high refresh or 1440p at 60-90 Hz, where its extra VRAM and DLSS both help it hold that resolution better than the raster numbers alone suggest. Neither is a 4K card.
Running an older PCIe 3.0 board?
Both cards are unaffected by PCIe generation in practice. The RTX 3060 runs a full PCIe 4.0 ×16 link, and the GTX 1660 Ti's PCIe 3.0 ×16 link is nowhere near saturated by a card at its performance level, even on an older motherboard.
Price history & dollars-per-frame
Raw price is only half the story. Cost per frame is the number that decides it, and it keeps moving as prices fall.
New price over time
Street price of new cards (USD)
The 6 GB vs 12 GB question
GTX 1660 Ti · 6 GB
- 192-bit bus and 288 GB/s bandwidth are surprisingly close to the RTX 3060's, so the memory subsystem itself isn't dramatically outclassed.
- 6 GB still opens most esports and older AAA titles at 1080p high settings.
- 6 GB is tight for 2026 AAA at ultra with texture packs, well before the GPU core itself runs out of steam.
- No Tensor cores means no local AI use beyond the most basic CPU-assisted workloads.
RTX 3060 · 12 GB
- 12 GB on the same 192-bit bus width as the 1660 Ti is effectively double the usable headroom at the same bus class.
- The extra capacity is what keeps this card relevant for 1440p, emulation, and local AI in 2026.
- It costs roughly 2.5x as much on the used market, a real consideration on a tight budget.
- Real ray tracing and DLSS are wasted if you're still gaming at 1080p on a budget monitor.
Emulation & memory bandwidth
The narrower bus is a real downgrade for emulation
Both cards share a 192-bit memory bus, an unusual coincidence given how far apart they are generationally, but the RTX 3060's higher 360 GB/s bandwidth and double the VRAM still give it a real edge in demanding, high-resolution Switch and PS3 emulation over the 1660 Ti's 288 GB/s and 6 GB.
Creator, AI & streaming
The newer card is quicker per task, but the older card's extra VRAM unlocks larger jobs it can't fit at all.
Streaming: the AV1 encoder is the differentiator
Neither card has AV1. The RTX 3060's Ampere-generation NVENC encoder is a step up in quality and efficiency from the GTX 1660 Ti's Turing encoder, though both are limited to H.264 and HEVC.
Overclocking & undervolting
The RTX 3060 undervolts especially well and has a bit more core overclocking room too. The GTX 1660 Ti's headroom is similar in percentage terms but starting from a much lower base, neither turns into a meaningfully different card through tuning alone.
Pros & cons
GTX 1660 Ti
- Cheapest card in this entire comparison on the used market
- Efficient for its era, close frames-per-watt to the newer 3060
- Enough for 1080p esports and older AAA
- About 57% slower on average across modern AAA
- No ray tracing hardware and no DLSS at all
- Only 6 GB VRAM, the tightest of any budget card here
RTX 3060
- About 57% faster with real ray tracing and DLSS
- 12 GB VRAM keeps it relevant well into 2026
- Meaningfully better for 1440p, AI, and emulation
- Costs roughly 2.5x as much on the used market
- Draws 50 W more and runs a bit hotter
- Still no AV1 encoder
Buying pitfalls
No ray tracing on the GTX 1660 Ti, at all
The RT toggle in a game's menu will either be greyed out or run at unplayable frame rates through pure software fallback. Despite being Turing architecture like the RTX 20-series, the GTX-branded 1660 Ti was built without RT or Tensor cores, don't assume 'Turing' means it has ray tracing hardware, it doesn't.
Same memory bus, very different card
Both run a 192-bit bus, which can make them look more similar on paper than they are. Capacity (6 GB vs 12 GB) and bandwidth (288 vs 360 GB/s) still favor the RTX 3060 substantially, and the GPU core itself is a full generation ahead on top of that.
GTX 1660 Ti vs RTX 3060: your questions answered
Large. It's about 57% faster on average, has double the VRAM, and adds real ray tracing and DLSS that the 1660 Ti doesn't have at all. This is a genuine generation-and-a-half gap, not a close call.
If your budget is truly capped around $100, the 1660 Ti still plays 1080p games. If you can stretch to $260, the RTX 3060 is a substantially better card in every way that matters, more speed, more VRAM, and real ray tracing and DLSS.
For 1080p esports and older or lighter AAA titles, yes, it still works. For current AAA at real settings it's showing its age hard, and it has no ray tracing or DLSS to lean on when things get demanding.
Yes, this is one of the upgrades on this site that's actually worth making. A 57% average gain plus double the VRAM and real ray tracing and DLSS justifies the cost if you're serious about gaming going forward.
Close to it. The GTX 1660 Ti has no Tensor cores at all, so it's not just slower at AI work, it's barely usable for it. The 3060's 12 GB and Tensor cores make it a genuinely capable budget AI card by comparison.
Both help. Its newer NVENC encoder and much larger VRAM buffer together handle bigger timelines and cleaner output than the 1660 Ti can manage.
A bit, yes. Despite being built on the same Turing architecture family as the RTX 20-series, the GTX-branded 1660 Ti has no RT or Tensor cores. It can use FSR or XeSS instead, since those work on any GPU.
Neither supports it. The RTX 3060 has DLSS 2 upscaling only, which doesn't add the interpolation latency that true frame generation does, and the 1660 Ti has no DLSS of any kind.
Less than you'd expect for the 1660 Ti, it's slow enough that most CPUs from the last several years keep it fed. The RTX 3060 is more demanding; use the pairing checker on this page, a Ryzen 5 5600 or Core i5-12400F and up is a good match.
The GTX 1660 Ti draws noticeably less, 120 W against the RTX 3060's 170 W, and runs a bit cooler and quieter as a result, despite the 3060 being close in frames-per-watt overall.
It covers the GTX 1660 Ti comfortably; the RTX 3060 wants a quality 550 W supply instead. Both use a single 8-pin power connector.
Not really, PCIe generation isn't a real concern for either card at this performance level. Most 1660 Ti models run around 210-230 mm and most 3060 models around 242 mm, so check your case's maximum GPU length instead.
No, for either. Mobile GTX 1660 Ti and mobile RTX 3060 variants are different, lower-power chips than their desktop namesakes. Never compare laptop numbers against desktop numbers, use a mobile-specific comparison instead.
No, not for modern AAA games. Both are 1080p cards; the RTX 3060 can stretch into light 1440p with DLSS, the 1660 Ti has no upscaling to help it reach further.
Yes, buy from a seller with a return window and test it. The RTX 3060 can often still be found new at close to its street price, worth it for the warranty if the budget allows.
A bit, yes. The RTX 3060's Ampere architecture has years of full Game Ready driver support ahead of it. The GTX 1660 Ti's Turing-GTX branch still gets updates but is further along in its support lifecycle, treat it as a shorter-term card.