GTX 1060 vs GTX 1660
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 1060
GeForce GTX 1660
avg, 1080p
The short answer
The GTX 1660 is clearly the better card, about 34% faster on average at the same 6 GB of VRAM and similar power draw, no real trade-off involved. The GTX 1060 only makes sense if it's meaningfully cheaper, since neither card supports ray tracing or DLSS and both are well past their prime for current AAA gaming. If you're spending real money on either, consider a modern budget card like the RTX 3050 or RX 6600 instead.
Two generations of Nvidia's budget mainstream line, three years apart. The GTX 1060 was 2016's default 1080p card, the GTX 1660 replaced it in 2019 with a newer Turing core at the same 6 GB and similar power draw. Neither has ray tracing or DLSS, this is a pure raster comparison between an old budget card and a slightly less old one.
What follows breaks down these two budget cards in full: spec sheet, ten-game frame rates with 1% lows, the ray tracing picture (neither card has the hardware), thermals, power, price history, and a plain verdict.
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.
The cheapest card in this whole comparison series, used or new, if every dollar counts.
34% more average fps for the same 6 GB and similar power draw makes this the easy pick if the price gap is small.
Neither is good at this, but the GTX 1660 is meaningfully less bad if you have no other option.
2nd-gen NVENC with HEVC B-frame support gives the GTX 1660 cleaner output at the same bitrate.
Power draw is essentially tied, but the newer Turing core is easier to keep cool and quiet at that same 120 W.
For games from the 1060's own era, the extra fps of the 1660 matters far less, buy the cheaper card.
Already own a GTX 1060? The GTX 1660 helps, but neither is a long-term answer.
Short answer: probably not
A 34% gain for the same VRAM and power draw is a real, worthwhile upgrade if the used price is low. It buys you smoother frame rates in current games without changing your PSU or case fit.
But neither card supports ray tracing or DLSS, and both are years past new-title driver priority. If you're spending money anyway, put it toward a modern budget card like the RTX 3050 or RX 6600 instead, you'll get the same or better fps plus a feature set that will actually matter over the next few years.
Full specification comparison
Every spec side by side. The highlighted figure wins each row.
| Specification | GTX 1060 | GTX 1660 |
|---|---|---|
| Architecture | ||
| GPU / architecture | GP106 · Pascal | TU116 · Turing |
| Process node | TSMC 16 nm | TSMC 12 nm ▲ |
| Release date | Jul 19, 2016 | Mar 14, 2019 |
| Launch MSRP | $249 | $219 ▲ |
| Compute | ||
| CUDA cores | 1,280 | 1,408 ▲ |
| RT cores | None, pre-RTX | None, cut-down Turing |
| Tensor cores | None | None |
| Base / boost clock | 1506 / 1708 MHz | 1530 / 1785 MHz ▲ |
| L2 cache | 1.5 MB | 1.5 MB |
| Memory | ||
| VRAM | 6 GB GDDR5 | 6 GB GDDR5 |
| Memory bus | 192-bit | 192-bit |
| Bandwidth | 192 GB/s | 192 GB/s |
| Power & form factor | ||
| TDP | 120 W | 120 W |
| Power connector | 1× 6-pin | 1× 8-pin |
| Suggested PSU | 400 W ▲ | 450 W |
| PCIe interface | 3.0 ×16 | 3.0 ×16 |
| Features | ||
| Upscaling | None, pre-DLSS | None, no Tensor cores |
| AV1 encode / decode | No | No |
| NVENC generation | 1st gen | 2nd gen, HEVC B-frames ▲ |
| Display outputs | 1×DL-DVI · HDMI 2.0b · 3×DP1.4 | 1×DL-DVI · HDMI 2.0b · 3×DP1.4 |
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 GTX 1660'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 GTX 1660's 8 GB VRAM does in a texture-heavy game.
Frametime: Hogwarts Legacy, 1080p Ultra
Lower and flatter is better · 16.7 ms = 60 fps
Counterintuitive but real: the slower GTX 1060 delivers a smoother line here because 12 GB holds the texture pool the 8 GB GTX 1660 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 GTX 1660 has a real edge.
Is frame generation "real" performance?
Neither card supports any form of frame generation, that technology, DLSS 3 and FSR 3 alike, arrived years after both of these architectures launched. Both can use basic FSR 1 upscaling since it works on any GPU, but it's a simple sharpening-based upscaler, not the temporal reconstruction of FSR 2/3 or DLSS.
With base frame rates this low in demanding modern games, frame generation wouldn't help much anyway, it needs a healthy base frame rate to feel good, and neither of these cards reliably hits that in current AAA titles at ultra settings.
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 1060, it will run the GTX 1660 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 1060: best kept to 1080p at 60-75 Hz on high to ultra settings in current games. GTX 1660: 1080p 60-100 Hz and workable at 1440p on lighter or older titles. Neither should be paired with a high-refresh 1440p or 4K panel expecting to fill it.
Running an older PCIe 3.0 board?
Both run PCIe 3.0 ×16. On a PCIe 4.0 or 5.0 motherboard they simply negotiate down to 3.0 speeds, with no performance penalty either way since neither card is remotely bandwidth-starved at ×16.
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 6 GB question
GTX 1060 · 6 GB
- 6 GB was generous for its era and still covers most 1080p titles at medium to high settings.
- 192 GB/s of bandwidth, identical to the GTX 1660, means it isn't memory-starved relative to its own core speed.
- The core itself is the real bottleneck now, extra memory bandwidth can't make up a 34% raw performance gap.
- No Tensor cores at all, so it can't run DLSS and is essentially unusable for GPU-accelerated AI work.
GTX 1660 · 6 GB
- Same 6 GB and 192 GB/s as the GTX 1060, so memory isn't the differentiator here, the newer Turing core is.
- Turing's improved memory compression stretches that same bandwidth further in practice than Pascal does.
- Still only 6 GB, a real limit in 2026's texture-heavy titles regardless of the architecture improvements.
- No Tensor or RT cores either, TU116 is a cut-down Turing die built specifically without them.
Emulation & memory bandwidth
The narrower bus is a real downgrade for emulation
Both handle Wii, PS2 and most Switch emulation comfortably. Push into demanding RPCS3 (PS3) titles or high internal resolutions and the GTX 1660's extra shader throughput gives it a clearer, more consistent edge than the identical memory bandwidth alone would suggest.
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
Both lack AV1 entirely, encode and decode. The GTX 1660's 2nd-generation NVENC handles HEVC B-frames the 1060's 1st-generation encoder can't, giving it modestly better quality per bitrate for streaming or recording.
Overclocking & undervolting
Both have real, old-school overclocking headroom, roughly 8 to 10% with a manual core and memory bump, since neither uses modern efficiency-limited boost algorithms. Neither has a meaningful undervolt-and-curve-optimizer path the way newer cards do.
Pros & cons
GTX 1060
- Cheapest card in this whole comparison series, used or new
- Identical 192 GB/s bandwidth to the newer 1660 despite the age gap
- Enough VRAM and grunt for older and lighter modern games at 1080p
- About 34% slower than the GTX 1660 on average
- No RT, no DLSS, and 1st-gen NVENC with no HEVC B-frames
GTX 1660
- About 34% faster than the GTX 1060 at the same 6 GB and TDP
- Newer Turing core with better memory compression and NVENC
- Still findable new-in-box in some markets, unlike the 1060
- No RT or Tensor cores either, it's a cut-down Turing die by design
- Still only 6 GB, the real ceiling for both cards in 2026
Buying pitfalls
GTX 1060 3GB vs 6GB, check carefully
There's also a cut-down GTX 1060 3GB with fewer CUDA cores and a smaller memory config that performs noticeably worse. Everything on this page refers to the common 6GB version, confirm the listing says 6 GB before buying used.
Both cards are past their practical shelf life
Neither supports ray tracing, DLSS, or FSR 3 frame generation, and both are years past new-game driver optimization priority. Treat either as a stopgap or a light/older-game build, not a card to buy expecting long-term headroom.
GTX 1060 vs GTX 1660: your questions answered
Yes, clearly, it's about 34% faster on average at the same 6 GB VRAM and similar power draw. There's no scenario where the older GTX 1060 outperforms it.
The GTX 1660 if the price gap is small, it's a straightforward 34% more performance for the same VRAM and power. Only choose the GTX 1060 if it's meaningfully cheaper and your budget is very tight.
It's usable for older or lighter games at 1080p medium to high settings, but it struggles with current AAA titles at ultra settings. It's a budget stopgap, not a card to plan around long-term.
It's dated but more usable than the 1060, handling 1080p high settings in most current games and medium to high in the heaviest titles. Like the 1060, treat it as a stopgap rather than a long-term build.
It's a real 34% gain for the same VRAM and power draw, worthwhile if the used price is low. But neither card supports ray tracing or DLSS, so if you're spending money anyway, a newer RTX 3050 or RX 6600 gets you those features too.
Neither is good at this. Both lack Tensor cores, so AI workloads fall back to slow general compute. The GTX 1660 is faster but still a poor choice, look at an RTX-series card instead for any serious AI work.
The GTX 1660, its 2nd-generation NVENC handles HEVC B-frames the 1060's older encoder lacks, giving cleaner output at the same bitrate. Neither supports AV1 in any form.
No. Both predate RTX, the GTX 1060 is Pascal and the GTX 1660 is a cut-down Turing die built specifically without RT or Tensor cores. Ray tracing modes simply won't run on either.
Neither supports DLSS, that needs Tensor cores. Both can use basic FSR 1 upscaling since it works on any GPU, but not the more advanced temporal FSR 2 or 3.
It's the opposite concern with cards this old, they're more likely to bottleneck a modern CPU than be bottlenecked by one. Even a budget current CPU like a Ryzen 5 5500 easily outpaces what either GPU can feed it.
Both draw around 120 W under gaming load, essentially identical. Noise depends more on the specific AIB cooler than the architecture at this power level.
A 400 W quality unit covers the GTX 1060, and 450 W covers the GTX 1660. Either fits comfortably in a basic 450-500 W budget build.
Typical GTX 1060 cards run around 250 mm and GTX 1660 cards around 229 mm, check your case's clearance. Both use PCIe 3.0 ×16 and run at full speed on any modern motherboard, no penalty either way.
No, not for any current game at real settings. Both are 1080p cards, full stop, 1440p is a stretch even on older or lighter titles.
Both are used-market cards at this point, new stock is essentially gone. Buy from a seller with a return window, avoid untested ex-mining cards, and expect to replace either within a couple of years if you play current games.
Both still receive basic security and compatibility drivers, but neither gets new feature work or the same Game Ready optimization priority as current cards. Budget for a replacement sooner rather than later if you play new releases.