GTX 1650 vs GTX 1050 Ti
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 1650
GeForce GTX 1050 Ti
avg, 1080p
The short answer
The GTX 1650 is clearly the faster card, about 40% ahead on average thanks to a newer architecture and more CUDA cores at a higher clock. Both cards carry the same 4 GB of GDDR memory on a 128-bit bus, and neither has ray tracing hardware, DLSS, or an AV1 encoder, this is an entry-level matchup with no bells or whistles on either side. The GTX 1050 Ti's only real edge is price on the used market. If both are within a few dollars of each other, buy the 1650. Only take the 1050 Ti if it's meaningfully cheaper and you're building on the tightest possible budget.
These two show up together because they occupy the exact same market slot three years apart, the cheapest new GPU NVIDIA sold at the time. The GTX 1050 Ti was that card in 2016, the GTX 1650 took over the role in 2019. Both target 1080p on a budget, both skip ray tracing hardware entirely, and both are now sold almost exclusively used.
Everything that follows details these two entry cards: specs, ten-game 1080p frame rates, the absent ray tracing story, power and thermals, price history, and a verdict for the tightest budgets.
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 1650's 40% lead matters most in competitive titles where every extra frame helps hit a stable 100+ fps.
More headroom across the board, the 1650 stays comfortably above the 1050 Ti in every game tested here.
Neither has Tensor Cores, so both are slow, but the 1650's faster CUDA throughput still finishes work sooner on the same 4 GB.
Turing's 7th-gen NVENC encodes noticeably cleaner video than the 1050 Ti's older Pascal encoder at the same bitrate.
Both skip the power connector entirely, but the 1650 gives more performance for the same 300 W PSU headroom.
If the 1050 Ti is meaningfully cheaper used, it still plays esports and older games fine, just with less headroom.
Already own a GTX 1050 Ti or GTX 1650? Neither is worth trading for the other.
Short answer: probably not
A 40% gain sounds like a lot, but in absolute terms it's the difference between 22 and 31 fps average, still well short of a smooth, modern 1080p experience. If you already own either card, the upgrade math doesn't favor a sidegrade to the other one.
The real jump from this tier is an RTX 3050 or RX 6600, both roughly double this pair's performance and add ray tracing, DLSS or FSR, and enough VRAM headroom to matter. Save toward one of those instead of shuffling between these two.
Full specification comparison
Every spec side by side. The highlighted figure wins each row.
| Specification | GTX 1650 | GTX 1050 Ti |
|---|---|---|
| Architecture | ||
| GPU / architecture | TU117 · Turing | GP107 · Pascal |
| Process node | TSMC 12 nm ▲ | Samsung 14 nm |
| Release date | Dec 12, 2019 (GDDR6) | Oct 25, 2016 |
| Launch MSRP | $159 | $139 ▲ |
| Compute | ||
| CUDA cores | 896 ▲ | 768 |
| RT cores | None | None |
| Tensor cores | None | None |
| Base / boost clock | 1410 / 1590 MHz ▲ | 1290 / 1392 MHz |
| L2 cache | 1 MB | 1 MB |
| Memory | ||
| VRAM | 4 GB GDDR6 ▲ | 4 GB GDDR5 |
| Memory bus | 128-bit | 128-bit |
| Bandwidth | 192 GB/s ▲ | 112 GB/s |
| Power & form factor | ||
| TDP | 75 W | 75 W |
| Power connector | None | None |
| Suggested PSU | 300 W | 300 W |
| PCIe interface | 3.0 ×16 | 3.0 ×16 |
| Features | ||
| Upscaling | FSR 2 (no DLSS) | FSR 2 (no DLSS) |
| Video encode (NVENC) | 7th gen · Turing ▲ | 6th gen · Pascal |
| AV1 encode (NVENC) | No | No |
| Display outputs | DP1.4 · HDMI2.0b · DL-DVI-D | DP1.4 · HDMI2.0b · DL-DVI-D |
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 1050 Ti'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 1050 Ti's 8 GB VRAM does in a texture-heavy game.
Frametime: Hogwarts Legacy, 1080p High
Lower and flatter is better · 16.7 ms = 60 fps
Counterintuitive but real: the slower GTX 1650 delivers a smoother line here because 12 GB holds the texture pool the 8 GB GTX 1050 Ti 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 1050 Ti has a real edge.
Is frame generation "real" performance?
Neither card supports DLSS Frame Generation, that needs Tensor Cores and an Optical Flow Accelerator, hardware that only arrived with RTX 40-series cards. The GTX 1650 has no Tensor Cores, and the GTX 1050 Ti predates Tensor Cores entirely, both are pre-RTX designs from the ground up.
AMD's FSR 3 Frame Generation runs through the game engine rather than fixed hardware, so it technically works on both, but it needs a reasonably smooth base frame rate to look right, something neither card reliably hits in demanding games. Treat frame generation as unavailable for this pairing.
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 1650, it will run the GTX 1050 Ti 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 1650: best at 1080p on medium to high settings, 60 to 90 Hz depending on the game. GTX 1050 Ti: 1080p on medium settings, generally capped closer to 60 Hz in anything demanding. Neither is a 1440p card outside of light or older titles.
Running an older PCIe 3.0 board?
Both cards run a full PCIe 3.0 x16 link, so motherboard generation is not a bottleneck for either one. Neither needs PCIe 4.0, and both run at full speed even in an older 3.0-only slot.
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 4 GB vs 4 GB question
GTX 1650 · 4 GB
- GDDR6 gives the same 4 GB more real bandwidth than GDDR5, so the pool is used more efficiently under load.
- Enough for esports titles and most games at 1080p medium settings without hitting a hard wall.
- Faster core means less pressure on the memory system per frame than the slower 1050 Ti.
- Still only 4 GB, modern AAA titles at high textures push past this on either card.
GTX 1050 Ti · 4 GB
- Same 4 GB capacity as the 1650, so it handles the same games at the same settings before running out of room.
- GDDR5 at 112 GB/s is nearly 40% less bandwidth than the 1650's GDDR6, a real bottleneck at higher internal resolutions.
- The slower core means it hits VRAM and bandwidth limits sooner, at lower settings, than the 1650 does.
- A decade-old memory standard, no benefit from any modern compression improvements built into newer memory controllers.
Emulation & memory bandwidth
The narrower bus is a real downgrade for emulation
Both cards cap out at 4 GB, so neither is built for pushing internal resolution scaling far in demanding emulators. The GTX 1650's 192 GB/s of bandwidth against the 1050 Ti's 112 GB/s matters more here than in native games, bandwidth-bound cores like RPCS3 or Switch emulation at 2x native and up will stutter on the 1050 Ti well before the 1650 struggles. For light emulation, PS2, GameCube, Wii, either card is fine.
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 an AV1 encoder, both top out at H.264 and HEVC through NVENC. The GTX 1650's 7th-generation Turing encoder is a real step up in quality per bitrate over the 1050 Ti's older 6th-generation Pascal encoder, so it streams and records a bit cleaner at the same settings even though it's already the faster card for gaming.
Overclocking & undervolting
Neither card has much headroom left at 75 W. The Pascal-based GTX 1050 Ti has a slight edge in classic Pascal overclocking behavior, but on a card this small the gain is only a few percent either way. Undervolting does more for noise and heat than overclocking does for frame rate on both.
Pros & cons
GTX 1650
- About 40% faster on average across modern titles
- GDDR6 memory gives real bandwidth headroom over the 1050 Ti's GDDR5
- Newer Turing NVENC encoder streams a bit cleaner
- Costs more, new or used
- Still no ray tracing, DLSS or AV1, same ceiling as the 1050 Ti
GTX 1050 Ti
- Cheapest card in this comparison, new or used
- Same 4 GB capacity as the 1650 at a lower price point
- No power connector, drops into almost any prebuilt or small case
- About 40% slower on average across modern titles
- Slower GDDR5 memory bandwidth limits it sooner at higher resolutions
Buying pitfalls
Which GTX 1650 are you buying?
There are two GTX 1650s: the original 2019 GDDR5 card and a late-2019 GDDR6 refresh with roughly 50% more memory bandwidth and noticeably better 1% lows. This page uses the GDDR6 version. Check the exact model name before buying, the GDDR5 version closes much of the gap to the 1050 Ti.
GTX 1050 vs GTX 1050 Ti, don't mix them up
The plain GTX 1050 is a separate, weaker card with 640 CUDA cores and only 2 GB of VRAM, not a naming variant of the Ti. This page covers the 4 GB GTX 1050 Ti only. Confirm the listing says 'Ti' and '4GB' before buying, the non-Ti loses to both cards on this page.
GTX 1650 vs GTX 1050 Ti: your questions answered
Yes, for gaming it's about 40% faster on average thanks to a newer architecture, more CUDA cores, and faster GDDR6 memory. Both share the same 4 GB capacity, neither has ray tracing or DLSS.
Buy the 1650 if the price difference is small, it's meaningfully faster for a similar cost. Buy the 1050 Ti only if it's noticeably cheaper used and you're building on the tightest possible budget.
For esports and older or lighter games at 1080p, yes, both remain usable. For current AAA titles, both struggle badly, this comparison's own numbers show sub-25fps averages in demanding games like Alan Wake 2.
Neither is a real choice for AI work, both lack Tensor Cores entirely and cap out at 4 GB of VRAM. If forced to pick, the GTX 1650's faster CUDA throughput finishes the same task sooner than the 1050 Ti.
The GTX 1650, its Turing NVENC encoder produces cleaner video at the same bitrate than the 1050 Ti's older Pascal encoder. Neither has an AV1 encoder or enough VRAM for serious editing work.
No. Both are pre-RTX designs with zero dedicated RT hardware. Ray tracing only runs through slow software fallback, effectively unplayable on either card.
Neither supports DLSS, that needs Tensor Cores and neither card has any. Both can use AMD's open FSR 2 upscaler, which works on any GPU regardless of brand.
Rarely. These are entry-level GPUs, so almost any CPU from the last decade keeps them fed at 1080p. Use the pairing checker on this page to confirm your specific CPU.
Both draw around 70 W and need no power connector at all, drawing everything from the PCIe slot. Any PC built in the last decade already has enough PSU headroom for either.
NVIDIA lists a 300 W system requirement for both cards. Any decent 400 to 450 W unit comfortably covers either with room to spare.
Both cards run a full PCIe 3.0 x16 link, unaffected by motherboard generation. Physical size is small on both, around 170mm for the 1650 and 145mm for a typical 1050 Ti, check your case's minimum clearance either way.
Yes, both had mobile variants, GTX 1650 Mobile and GTX 1050 Ti Mobile, running at lower clocks and power limits than the desktop cards covered here. Never compare laptop and desktop numbers directly.
No, not for modern games. Both are 1080p cards at best, and even that requires dialing back settings in demanding titles. Treat 4K as unrealistic for this pairing.
Both are essentially used-market-only cards in 2026. Buy from a seller with a return window, avoid untested ex-mining cards, and expect to pay in the $40 to $85 range depending on which card and condition.
NVIDIA continues Game Ready driver support for both Turing and Pascal architectures, so neither is at immediate risk of losing updates, though don't expect either to be a priority for new game optimizations going forward.
Confirm you're getting the GDDR6 GTX 1650 (not the slower original GDDR5 version) and the 4GB GTX 1050 Ti (not the weaker plain GTX 1050 with 2 GB). Both names cover more than one real variant with a meaningful performance gap.