Pick two to four cards and see them side by side across price, performance, memory, power, technologies and physical size. Every comparable row marks a winner; rows where the cards tie, or where CompareYourGPU has no data, say so instead of guessing.
Pick two to four cards. 2 of 4 selected.
Scroll the table horizontally on narrow screens. The specification column stays fixed.
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| Launch MSRPEssentially tied | $599 | $599 |
| Current priceCompareYourGPU does not publish a price it has not verified. | $1,029 | Best in this row: $988 |
| Performance per $100Performance index at 1440p per $100 of verified current price. | 3.6 | Best in this row: 6.4 |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| CompareYourGPU estimate, 100 = fastest card here. | 42 | Best in this row: 70 |
| CompareYourGPU estimate, 100 = fastest card here. | 37 | Best in this row: 63 |
| CompareYourGPU estimate, 100 = fastest card here. | 31 | Best in this row: 57 |
| CompareYourGPU estimate, 100 = fastest card here. | 31 | Best in this row: 53 |
| Measured suite average at 1440pNot compared: data missing for at least one cardThird-party geometric mean across their game suite, run with vendor upscaling at Quality. A different test from the estimated row below, not a check on it. | Not tested | 88.8 FPS (ComputerBase) |
| Estimated FPS at 1440pModelled estimate for a typical modern AAA game at High settings with no upscaling. Not a measurement. | 65-95 FPS | Best in this row: 115-165 FPS |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| 8 GB | Best in this row: 16 GB | |
| GDDR6X | GDDR6 | |
| Essentially tied | 256-bit | 256-bit |
| 608 GB/s | Best in this row: 640 GB/s | |
| Memory speed | 19 Gbps | Best in this row: 20 Gbps |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| Architecture | Ampere | RDNA 4 |
| GPU die | GA104 | Navi 48 |
| Only comparable within one manufacturer. | Best in this row: 6,144 | 4,096 |
| Not compared: data missing for at least one card | Data unavailable | 64 |
| 48 | Best in this row: 64 | |
| Best in this row: 192 | 128 | |
| Not compared: data missing for at least one card | 1,580 MHz | Data unavailable |
| 1,770 MHz | 2,970 MHz | |
| Game clockNot compared: data missing for at least one cardAMD only. | Data unavailable | 2,400 MHz |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| Best in this row: 290 W | 304 W | |
| Essentially tiedVendor figure where published, otherwise board power plus 300 W of headroom. | 750 W | 750 W |
| Performance per 100 W | 12.8 | Best in this row: 20.7 |
| Power connectors | 2x 8-pin | 2x 8-pin |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| 2nd-generation RT cores | 3rd-generation ray accelerators | |
| DLSS 4 Super Resolution, DLSS Ray Reconstruction, FSR 3.1, XeSS | FSR 4, FSR 3.1, XeSS | |
| FSR 3.1 Frame Generation (vendor-agnostic, game-dependent) | FSR 3.1 Frame Generation, AMD Fluid Motion Frames 2 (driver-level) | |
| No | Yes | |
| Video encoder | 7th-generation NVENC with HEVC and H.264 encode. No AV1 encode. | AMD Radeon media engine with AV1, HEVC and H.264 encode |
| Video decoder | 5th-generation NVDEC with AV1, HEVC and H.264 decode | AV1, HEVC and H.264 decode |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| DisplayPort | 3x 1.4a | 3x 2.1a |
| HDMI | 1x 2.1 | 1x 2.1b |
| USB-C | None | None |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| LengthNot compared: data missing for at least one card | 267 mm | Data unavailable |
| Height from bracketNot compared: data missing for at least one card | 112 mm | Data unavailable |
| Not compared: data missing for at least one card | 2 slots | Data unavailable |
| Specification | RTX 3070 TiNVIDIA | RX 9070 XTAMD |
|---|---|---|
| Launch date | June 10, 2021 | March 6, 2025 |
| Generation | RTX 30 | RX 9000 |
| Availability | Out of production | In production |
Each category is decided by the same numbers used everywhere else on the site. A category with no clear leader says so rather than picking one.
Performance at 1440p
Performance index 63
Ray tracing
Ray tracing index 53
Price
$988
Performance per dollar
6.4 index points per $100
VRAM
16 GB
Memory bandwidth
640 GB/s
Efficiency
20.7 index points per 100 W
Lowest power draw
290 W board power
All six axes are normalised to a 0-100 scale so they can share one chart. Performance and ray tracing are CompareYourGPU’s estimated indices; value, VRAM, efficiency and features are each card’s standing relative to every other card tracked here.
| Criterion | RTX 3070 Ti | RX 9070 XT |
|---|---|---|
| Performance | 37 | 63 |
| Ray tracing | 31 | 53 |
| Value | 28 | 65 |
| VRAM | 36 | 80 |
| Efficiency | 4 | 67 |
| Features | 60 | 100 |
CompareYourGPU estimate aggregated from public review coverage, not a measurement.
Modelled from the performance index. Not a measured benchmark. Real results vary widely by title.
Capacity alone is not the whole story: memory bus width and bandwidth decide how much of it the GPU can actually use.
A card with a verified current price is shown at that price, a dated snapshot; a card without one is shown at launch MSRP. Street prices move, so check the retailer.
Based on the specifications, estimated performance indices and prices on record for these cards. How verdicts are calculated.
Highest estimated performance index at 1440p of the cards compared here, at 63. It is the right answer if frame rate is the only thing you are optimising for.
Delivers the most performance per dollar of the cards compared, at $988. Choose this if the difference in price matters more to you than the difference in frame rate.
Strongest ray-traced performance here, with an estimated ray tracing index of 53 at 1440p. It also carries FSR 4, which is how ray tracing becomes practical.
Carries 16 GB, the most headroom of the cards compared at 1440p. Memory is usually what dates a card before raw speed does.
Best performance per watt of the cards compared, drawing 304 W. That means less heat, quieter fans and less pressure on your power supply.
Best balance of speed, price and memory of the cards compared: 50% performance at 1440p, 30% performance per dollar, 20% memory headroom. Not necessarily the fastest.
These matchups have their own pages with additional editorial context and a comparison-specific FAQ.
$986 vs $814
The closest thing to a straight fight in the mid-range: two cards launched days apart at almost the same price, with opposite strengths. AMD brings 16 GB of memory, NVIDIA brings stronger ray tracing and the wider upscaling catalogue.
$1,510 vs $988
Both manufacturers' strongest mid-to-high offering, separated by $150 of launch MSRP. Both carry 16 GB. The question is whether NVIDIA's ray tracing and DLSS advantage is worth the premium over AMD's rasterisation-per-dollar.
$906 vs $646
The two 16 GB cards under $450. Both pair a modest graphics processor with a generous memory pool on a narrow 128-bit bus, and both are aimed squarely at 1080p and entry-level 1440p.
$906 vs $634
Same graphics processor, same clocks, same bandwidth, half the memory, $50 apart. This is the clearest example of a decision that specification sheets make look trivial and real-world use does not.
$6,592 vs $1,973
Twice the price for a large but not proportional performance gain, plus double the memory and a 215 W increase in board power. The question is whether the top of the range is a product or a statement.
$435 vs $383
The two cheapest current-generation cards, taking opposite approaches. Intel offers 12 GB and a wider memory bus; NVIDIA offers the DLSS ecosystem and lower power draw at the same price.
$502 vs $435
A $50 price gap between NVIDIA's entry 1080p card and Intel's memory-heavy alternative. One has the stronger software ecosystem, the other has 50 percent more VRAM.
$988 vs $1,973
A $400 gap between AMD's fastest RDNA 4 card and NVIDIA's second-tier Blackwell. Worth examining because the rasterisation gap is much smaller than the price gap, and the ray tracing gap is much larger.
$599 vs $499
Two previous-generation cards from opposite camps, and the question most second-hand and clearance shoppers face. NVIDIA's card is the faster of the two and has the better ray tracing and upscaling; AMD's launched $100 cheaper and carries 4 GB more memory.
$299 vs $386
The two entry-level cards of the last generation, both with 8 GB and both aimed at 1080p. They are level on estimated rasterisation performance, so the decision rests on power draw, ray tracing and upscaling rather than raw frame rates.
$986 vs $599
A question for anyone who already owns the older card: is the new one worth the upgrade? On estimated performance the two are level, and the RTX 5070 launched $50 cheaper. What separates them is the generation of features and how each is likely to age.
$1,599 vs $1,973
The old flagship against the new upper tier. The RTX 4090 is still faster and has 8 GB more memory, but it launched $600 higher, draws 90 W more and is out of production. The RTX 5080 gives up a little speed for a newer feature set and a current warranty.
$1,510 vs $1,973
Two cards from the same generation with the same 16 GB, $250 apart at launch. The RTX 5080 is the faster, but the gap is smaller in performance than in price, which is the whole of the decision.
$1,457 vs $999
Two cards at the same $999 launch price that take opposite approaches. AMD's flagship of its generation offers 24 GB of memory and level rasterisation; NVIDIA's offers far stronger ray tracing and DLSS. Both are discontinued, so condition and price matter.
$646 vs $561
One chip sold in two configurations $50 apart at launch, and the clearest example of why memory should be checked before price. The cards run at similar speeds in light games, and the larger one pulls away wherever memory becomes the limit.
$502 vs $561
Two 8 GB cards at an identical $299 launch price, from the two main camps. AMD's card is a little faster on paper and NVIDIA's has better ray tracing and a larger upscaling ecosystem. Both are best seen as 1080p cards.
$399 vs $634
A direct generational step at almost the same price. The newer card is clearly faster, uses GDDR7 and gains Multi Frame Generation, while keeping the same 8 GB and 128-bit bus. If you are choosing between the two on a shelf, this is an easy call.
$435 vs $352
Intel's two Battlemage cards, $30 apart at launch. The Arc B580 has more of everything: more memory, a wider bus and noticeably higher speed. For a small difference in price, it is the more convincing buy of the pair.
$699 vs $599
The old flagship against a newer mid-range card. They are level on estimated performance, but the RTX 4070 does it with 120 W less power, 2 GB more memory and frame generation. It is the classic case where the newer, smaller card is the better purchase.
$1,599 vs $6,592
Two generations of the fastest consumer card. The RTX 5090 is the clear winner on every measure, but it launched $400 higher and draws 125 W more. For most players the question is whether the extra speed and memory justify the premium.
$988 vs $1,457
AMD's newest upper mid-range card against its previous flagship. The older RX 7900 XTX is still the faster at 4K and carries 24 GB, but it launched $400 higher and lacks FSR 4. The RX 9070 XT closes most of the gap for a much lower price.
$1,411 vs $899
Two upper-mid-range cards of the last generation that trade blows in a familiar way. They are level in rasterisation, NVIDIA's has far better ray tracing and DLSS, and AMD's carries 4 GB more memory. Both are discontinued, so condition and price decide more than the specification.
$499 vs $399
A rare case where the older card is simply faster. The RTX 3070 outpaces the newer RTX 4060 Ti 8GB thanks to a wider memory bus, though the newer card is cheaper to run and adds DLSS frame generation. Both have the same 8 GB, so the limit is the same.
$555 vs $299
The older card with more memory against the newer card with better efficiency and features. They are level on estimated speed, which makes the decision about memory against power draw and frame generation. Both are discontinued.
$634 vs $561
Two 8 GB cards from the current generation, $80 apart at launch. The RTX 5060 Ti is faster and has the stronger ray tracing and feature set; the RX 9060 XT costs noticeably less. It is a question of whether the extra speed justifies the premium.
$986 vs $906
Speed against memory in the same generation. The RTX 5070 is far faster, but it has 12 GB where the RTX 5060 Ti 16GB has 16 GB, and the question is how much each matters to you.
$814 vs $988
AMD's two RDNA 4 upper-mid-range cards, with the same 16 GB and only $50 apart at launch. The XT is noticeably faster, and the plain card is considerably more efficient. For a small price gap, the choice is about power as much as speed.
$999 vs $1,510
The previous generation's upper tier against the current generation's step below it. The two are close in speed, and the newer card launched $250 cheaper with the newer feature set. It is a clear example of a generation lowering the price of a given level of performance.
$499 vs $814
AMD's previous upper mid-range card against its current one. Both have 16 GB, and the newer card is a large step faster, draws less power and supports FSR 4 for only $50 more at launch. For anyone choosing between them new, it is not a close decision.
$435 vs $561
Two budget cards that make opposite bets. Intel's gives 12 GB of memory for $50 less; AMD's gives more speed from a more mature platform but only 8 GB. It is the budget version of the speed-against-memory question.
