OVERALL
ALTERNATIVES
SPECS
VS AVERAGE
REVIEWS
Scored using Graphics cards Methodology v3.0
Graphics cards Methodology v3.0
Mathematical & Unbiased Scoring
Products in this category are scored under Frontum Graphics cards Methodology v3.0: the technical score from the specifications is the base, and independent expert and video reviews (up to ±0.5 points), owner opinions (up to ±0.3) and best-of lists (up to +0.3) move it up or down.
Explore Graphics cards Methodology v3.0

NVIDIA GeForce RTX 5090 review: 117 specs compared

Avg. price:~£1,850VRAM:32 GBMemory bus width:512 bitThermal Design Power (TDP):575 W

NVIDIA GeForce RTX 5090 Review: Compare 117 technical specifications, 59 expert and video reviews and 23 best-of lists to see how this product ranks in the Graphics cards category and whether it is worth buying.

NVIDIA GeForce RTX 5090
8.8/10

Overall score

Score
8.8/ 10
✕
What it is
Our main score for this model, 1–10. It combines specs, expert reviews, best-of lists and owners.
How we count it
Parts: Technical score 40%, Reviews score 25%, List score 25%, Sentiment score 10%
Marks: Each part gets a mark out of 10. For reviews, lists and owners, a typical model in this category gets about 8 and the best one gets 10.
No data yet: A part with no data shows "?". It counts as the mark of a weaker model in this category (4 in 5 models with data score higher). This way, models with reviews, lists and owners usually come out ahead.
Score distribution
▼
<55-66-77-88-99-10
★ Better than 99% of models
Score components

40%

9.1

Technical score

25%

8.3

Reviews score

25%

9.1

List score

10%

8.1

Sentiment score

★Better than 99% of graphics cards
6.5

Value for money

Value for money
6.5/ 10
✕
What it is
An indicator that combines the graphics card's overall rating with its cost.
When it matters
When you are looking for a graphics card that offers a strong balance of performance, features, and price.
Score distribution
▼
<55-66-77-88-99-10
▼ Worse than 86% of models
Score components

70%

8.8

Overall score

30%

1.0

Price

▼Worse than 86%
9.1

Technical score

Technical score
9.1/ 10
✕
What it is
How good the specs are compared with the other models in this category, 1–10.
How we count it
Areas: We rate the specs in each area against the other models in this category.
Weighting: The areas that matter most to buyers count more. The shares are shown under Score components.
Share: The technical score is 40% of the overall score.
Score distribution
▼
<55-66-77-88-99-10
★ Better than 90% of models
Score components

44%

9.6

Performance

24%

9.5

Memory

12%

6.1

Power

11%

10

Features

5%

7.5

Design

4%

10

Connectivity

★Better than 90%
8.3

Reviews score

Reviews score
8.3/ 10
✕
What it is
Average grade from independent test sites, magazines and video reviewers, 1–10.
How we count it
Grades: Each review's own score is converted to 1–10; without one, we grade the review ourselves. Big test sites count more, short reviews half as much as full tests. A video review counts half.
Score: The average grade is 80% of the review mark; the number of reviews is the other 20% (1 review = 5, the typical number in this category = 8, the most-reviewed model = 10).
Enough data: With few reviews, the average grade is pulled toward the typical grade of this category. No reviews yet: 7.6 for now.
Share: Reviews count 25% of the overall score.
Score distribution
▼
<55-66-77-88-99-10
▲ Better than 77% of models

80%

7.9

Average grade

20%

9.7

Popularity

Grades by review (59)
35 reviews: 20 with a printed score, 15 estimated from the review. 24 video reviews, each counting half as much as a written test. Reviews count 25% of the overall score.
User ratings collected online are not part of this score.
▲Better than 77%
9.1

List score

List score
9.1/ 10
✕
What it is
How often independent test sites, magazines and video reviewers put this model in their best-of lists for this category.
How we count it
Points: The model in the most lists of this category gets 10; the others get 8 to 10 in proportion. In no list yet: 7.5. Lists count 25% of the overall score.
Enough data: Every list counts, even a single one. With fewer than 5 best-of lists in the category, this part is left out.
Lists with this model (23)
List score: in 23 of 85 lists (27%)
★Better than 99%
  • nvidia-geforce-rtx-5090
  • nvidia-geforce-rtx-5090
  • nvidia-geforce-rtx-5090
  • nvidia-geforce-rtx-5090
  • nvidia-geforce-rtx-5090
  • nvidia-geforce-rtx-5090
  • nvidia-geforce-rtx-5090
nvidia-geforce-rtx-5090
nvidia-geforce-rtx-5090
nvidia-geforce-rtx-5090
nvidia-geforce-rtx-5090
nvidia-geforce-rtx-5090
nvidia-geforce-rtx-5090
nvidia-geforce-rtx-5090
Where to buy
*As an Amazon Associate, we earn from qualifying purchases.
Best rankings
Available: ranking among products currently available (including other versions of this product).All: ranking among all products in the database.

What reviewers say about the NVIDIA GeForce RTX 5090

Based on 29 · 24

Reviewers praise its cooling, compact size and AI muscle, but criticise the steep price, heavy power draw and connector worries.
Praised
Mixed views
Criticised
Pick a topic to see who said what

What owners say about the NVIDIA GeForce RTX 5090

Based on 241 owners sharing their experience in online comments

69%positive11%mixed20%negative

Owners praise its exceptional gaming speed, while high power draw, heat, and uneven performance at demanding settings temper the excitement.

What owners dislike

  • Owners say it misses high frame-rate targets, such as 240 or 300 fps at 4K or 1440p, and some games drop below 50 fps without frame generation.6
  • Owners report it pulls about 570 to 600 watts under full load, pushing PC draw to 700 to 800 watts and raising electricity bills.6
  • Owners report crashes and stutters, some fixed by removing GeForce Experience, and a recent update that enabled shared memory and slowed games.5
  • Owners say 4K with path tracing or ultrawide 5K-class resolutions need DLSS upscaling, and some games still drop to about 50 fps.4
  • Owners report distracting ghosting, blur and artifacts with DLSS and 4x frame generation, leading some to turn these features off.4
  • Owners say the native gain over a 5080 is small for the extra cost, and one claims a 4090 is faster.4
  • Owners say it runs hot, reaching 75 to 89°C in some builds, and gives off enough heat to noticeably warm the room.3
  • Owners say frame generation adds noticeable input lag, making mouse aiming feel sluggish or rubbery, especially in Stalker 2.3

Points mentioned by at least 3 owners, most mentioned first.

Loading category ranking...

Technical Specifications of graphics card NVIDIA GeForce RTX 5090

Technical score
Technical score
9.1/ 10
✕
What it is
How good the specs are compared with the other models in this category, 1–10.
How we count it
Areas: We rate the specs in each area against the other models in this category.
Weighting: The areas that matter most to buyers count more. The shares are shown under Score components.
Share: The technical score is 40% of the overall score.
Score distribution
▼
<55-66-77-88-99-10
★ Better than 90% of models
Score components:

44%

9.6

Performance

24%

9.5

Memory

12%

6.1

Power

11%

10

Features

5%

7.5

Design

4%

10

Connectivity

9.1
Reviews score
8.3
List score
9.1
Value for money
Value for money
6.5/ 10
✕
What it is
An indicator that combines the graphics card's overall rating with its cost.
When it matters
When you are looking for a graphics card that offers a strong balance of performance, features, and price.
Score distribution
▼
<55-66-77-88-99-10
▼ Worse than 86% of models
Score components:

70%

8.8

Overall score

30%

1.0

Price

6.5
3DMark Time Spy benchmark score
43,370 points
Market range: <4.1k points – >30.1k points
This graphics card scores 43,370 points in 3DMark Time Spy, which is higher than 98.5% of graphics cards and equal to 0.3% of graphics cards.
3DMark Port Royal score
38,000 points
Market range: <0 points – >22.4k points
This graphics card scores 38,000 points in 3DMark Port Royal, which is higher than 99.3% of graphics cards and equal to 0.1% of graphics cards.
PassMark (G3D) result
38,850 points
Market range: <7.7k points – >36k points
This graphics card scores 38,850 points in PassMark G3D, which is higher than 96.2% of graphics cards and equal to 0.1% of graphics cards.
PassMark (DirectCompute) result
24,712 points
Market range: <3.1k points – >26.3k points
This graphics card scores 24,712 points in PassMark DirectCompute, which is higher than 86.8% of graphics cards.
FP32 performance
104.8 TFLOPS
Market range: <3.94 TFLOPS – >56.3 TFLOPS
This graphics card delivers 104.8 TFLOPS FP32 performance, which is higher than that of 97.3% of graphics cards and equal to that of 1.1% of graphics cards.
Show more
What reviewers say
VRAM
32 GB
Market range: <4 GB – >16 GB
This graphics card has 32 GB of VRAM, which is more than 96.6% of graphics cards and equal to 3.1% of graphics cards.
Memory type
GDDR7
This graphics card uses GDDR7 memory, which is newer than on 79.6% of graphics cards and equal to 20.4% of graphics cards.
Memory bus width
512 bit
Market range: <128 bit – >384 bit
This graphics card uses a 512 bit memory bus, which is wider than that of 96.7% of graphics cards and equal to that of 2.8% of graphics cards.
Maximum memory bandwidth
1,792 GB/s
Market range: <168 GB/s – >960 GB/s
This graphics card reaches 1,792 GB/s memory bandwidth, which is higher than that of 97.9% of graphics cards and equal to that of 2.1% of graphics cards.
Effective memory speed
28,000 MHz
Market range: <7k MHz – >28k MHz
This graphics card reaches an effective memory speed of 28,000 MHz, which is faster than that of 78.9% of graphics cards and equal to that of 16.8% of graphics cards.
Show more
What reviewers say
PCI Express (PCIe) version
5
This graphics card supports PCIe 5, which is newer than on 73% of graphics cards and equal to 27% of graphics cards.
PCIe lanes
x16
This graphics card uses x16 PCIe lanes, which is more than 24.1% of graphics cards and equal to 75.9% of graphics cards.
DirectX version
12 Ultimate
This graphics card supports DirectX 12 Ultimate, which is more advanced than on 22.8% of graphics cards and equal to 77.2% of graphics cards.
Vulkan version
1.4
This graphics card supports Vulkan 1.4, which is more advanced than on 20.8% of graphics cards and equal to 79.2% of graphics cards.
OpenGL version
4.6
This graphics card supports OpenGL 4.6, which is more advanced than on 2% of graphics cards and equal to 98% of graphics cards.
Show more
What reviewers say
Max displays supported
4
Market range: <4 – >4
This graphics card supports up to 4 displays, which is more than 9% of graphics cards and equal to 88.2% of graphics cards.
Max digital resolution
7680x4320
This graphics card supports a maximum digital resolution of 7680x4320, which is higher than that of 9.2% of graphics cards and equal to that of 90.9% of graphics cards.
DisplayPort outputs
3
Market range: <1 – >3
This graphics card offers 3 DisplayPort outputs, which is more than 21.7% of graphics cards and equal to 76% of graphics cards.
DisplayPort version
2.1b
This graphics card supports DisplayPort 2.1b, which is more advanced than on 77.8% of graphics cards and equal to 22.2% of graphics cards.
DisplayPort link rates
80 Gbps
Market range: <5.4 Gbps – >80 Gbps
This graphics card supports DisplayPort link rates up to 80 Gbps, which is faster than on 84.9% of graphics cards and equal to 14.9% of graphics cards.
Show more
Thermal Design Power (TDP)
575 W
Market range: <86 W – >360 W
This graphics card has a TDP of 575 W, which is higher than that of 97.4% of graphics cards and equal to that of 2.5% of graphics cards.
Power consumption while under peak load
575 W
Market range: <106 W – >365 W
This graphics card draws 575 W under peak load, which is higher than 97% of graphics cards and equal to 1.6% of graphics cards.
Recommended PSU wattage
1,000 W
Market range: <450 W – >850 W
This graphics card recommends a 1,000 W PSU, which is higher than that of 95.9% of graphics cards and equal to that of 3.8% of graphics cards.
Board power limit
600 W
Market range: <107 W – >400 W
This graphics card has a board power limit of 600 W, which is higher than that of 98.1% of graphics cards and equal to that of 1.8% of graphics cards.
PCIe power spec
575 W
Show more
What reviewers say
Size
304x 137x 2 mm
Length
304 mm
Market range: <200 mm – >337 mm
This graphics card is 304 mm long, which is longer than 63% of graphics cards and equal in length to 1.8% of graphics cards.
Height
137 mm
Market range: <109 mm – >146 mm
This graphics card is 137 mm tall, which is taller than 74.1% of graphics cards and equal in height to 1.8% of graphics cards.
Slot width
2 slot/s
Market range: <2 slot/s – >3.25 slot/s
This graphics card occupies 2 slot/s, which is slimmer than 52.4% of graphics cards and equal in width to 43.4% of graphics cards.
Weight
2000 g
Market range: <541 g – >2k g
This graphics card weighs 2000 g, which is heavier than 89.1% of Graphics cards and equal to 2.2% of Graphics cards.
Show more
What reviewers say

NVIDIA GeForce RTX 5090 vs the average graphics card

246810TotalPerformanceMemoryFeaturesConnectivityPowerDesign
  • 130 more compute units
    This graphics card has more compute units than the average graphics card (170 vs 40). The average graphics card has 40 compute units.
    170 vs 40
  • 130 more ray tracing cores
    This graphics card has more ray tracing cores than the average graphics card (170 vs 40). The average graphics card has 40 ray tracing cores.
    170 vs 40
  • 4x higher memory bandwidth
    This graphics card has a higher memory bandwidth than the average graphics card (1,792 GB/s vs 448 GB/s). The average graphics card has a memory bandwidth of 448 GB/s.
    1,792 GB/s vs 448 GB/s
  • 20 GB more VRAM
    This graphics card has more VRAM than the average graphics card (32 GB vs 12 GB). The average graphics card has 12 GB VRAM.
    32 GB vs 12 GB
  • 320 bit wider memory bus
    This graphics card has a wider memory bus than the average graphics card (512 bit vs 192 bit). The average graphics card has a memory bus width of 192 bit.
    512 bit vs 192 bit
  • 4.33x higher texture rate
    This graphics card has a higher texture rate than the average graphics card (1,636.8 GTexel/s vs 378 GTexel/s). The average graphics card has a texture rate of 378 GTexel/s.
    1,636.8 GTexel/s vs 378 GTexel/s
  • 496 more TMUs
    This graphics card has more TMUs than the average graphics card (680 vs 184). The average graphics card has 184 TMUs.
    680 vs 184
  • 112 more ROPs
    This graphics card has more ROPs than the average graphics card (176 vs 64). The average graphics card has 64 ROPs.
    176 vs 64
  • 19.7% higher list score
    List score
    9.1/ 10
    ✕
    What it is
    How often independent test sites, magazines and video reviewers put this model in their best-of lists for this category.
    How we count it
    Points: The model in the most lists of this category gets 10; the others get 8 to 10 in proportion. In no list yet: 7.5. Lists count 25% of the overall score.
    Enough data: Every list counts, even a single one. With fewer than 5 best-of lists in the category, this part is left out.
    Lists with this model (23)
    List score: in 23 of 85 lists (27%)
    9.1 vs 7.6
  • 15.3% higher review score
    Reviews score
    8.3/ 10
    ✕
    What it is
    Average grade from independent test sites, magazines and video reviewers, 1–10.
    How we count it
    Grades: Each review's own score is converted to 1–10; without one, we grade the review ourselves. Big test sites count more, short reviews half as much as full tests. A video review counts half.
    Score: The average grade is 80% of the review mark; the number of reviews is the other 20% (1 review = 5, the typical number in this category = 8, the most-reviewed model = 10).
    Enough data: With few reviews, the average grade is pulled toward the typical grade of this category. No reviews yet: 7.6 for now.
    Share: Reviews count 25% of the overall score.
    Score distribution
    ▼
    <55-66-77-88-99-10
    ▲ Better than 77% of models

    80%

    7.9

    Average grade

    20%

    9.7

    Popularity

    Grades by review (59)
    35 reviews: 20 with a printed score, 15 estimated from the review. 24 video reviews, each counting half as much as a written test. Reviews count 25% of the overall score.
    User ratings collected online are not part of this score.
    8.3 vs 7.2
  • 130 more compute units
    This graphics card has more compute units than the average graphics card (170 vs 40). The average graphics card has 40 compute units.
    170 vs 40
  • 130 more ray tracing cores
    This graphics card has more ray tracing cores than the average graphics card (170 vs 40). The average graphics card has 40 ray tracing cores.
    170 vs 40
  • 4.33x higher texture rate
    This graphics card has a higher texture rate than the average graphics card (1,636.8 GTexel/s vs 378 GTexel/s). The average graphics card has a texture rate of 378 GTexel/s.
    1,636.8 GTexel/s vs 378 GTexel/s
  • 496 more TMUs
    This graphics card has more TMUs than the average graphics card (680 vs 184). The average graphics card has 184 TMUs.
    680 vs 184
  • 112 more ROPs
    This graphics card has more ROPs than the average graphics card (176 vs 64). The average graphics card has 64 ROPs.
    176 vs 64
  • 536 more AI cores
    This graphics card has more AI cores than the average graphics card (680 vs 144). The average graphics card has 144 AI cores.
    680 vs 144
  • 2.63x higher pixel rate
    This graphics card has a higher pixel rate than the average graphics card (424.2 GPixel/s vs 161.3 GPixel/s). The average graphics card has a pixel rate of 161.3 GPixel/s.
    424.2 GPixel/s vs 161.3 GPixel/s
  • 4.64x higher FP32 performance
    This graphics card has a higher FP32 performance than the average graphics card (104.8 TFLOPS vs 22.6 TFLOPS). The average graphics card has FP32 performance of 22.6 TFLOPS.
    104.8 TFLOPS vs 22.6 TFLOPS
  • 4.15x higher ray tracing score
    This graphics card has a higher ray tracing score than the average graphics card (38,000 points vs 9,148 points). The average graphics card has a ray tracing score of 9,148 points.
    38,000 points vs 9,148 points
  • 2.93x higher gaming score
    This graphics card has a higher gaming score than the average graphics card (43,370 points vs 14,779 points). The average graphics card has a gaming score of 14,779 points.
    43,370 points vs 14,779 points
  • 71.6% higher PassMark score
    This graphics card has a higher PassMark score than the average graphics card (38,850 points vs 22,638 points). The average graphics card has a PassMark score of 22,638 points.
    38,850 points vs 22,638 points
  • 2 years newer
    This graphics card was released more recently than the average graphics card (2025 vs 2023).
    January 2025
    2025 vs 2023
  • 17,408 more FP32 units
    This graphics card has more FP32 units than the average graphics card (21,760 vs 4,352). The average graphics card has 4,352 FP32 units.
    21,760 vs 4,352
  • 3.42x higher FP64 performance
    This graphics card has a higher FP64 performance than the average graphics card (1.6 TFLOPS vs 0.48 TFLOPS). The average graphics card has FP64 performance of 0.48 TFLOPS.
    1.64 TFLOPS vs 0.48 TFLOPS
  • 5.9x higher BF16 performance
    This graphics card has a higher BF16 performance than the average graphics card (209.5 TFLOPS vs 35.48 TFLOPS). The average graphics card has BF16 performance of 35.48 TFLOPS.
    209.5 TFLOPS vs 35.48 TFLOPS
  • 3.55x higher FP16 performance
    This graphics card has a higher FP16 performance than the average graphics card (104.8 TFLOPS vs 29.5 TFLOPS). The average graphics card has FP16 performance of 29.5 TFLOPS.
    104.8 TFLOPS vs 29.5 TFLOPS
  • 5.1% higher base clock speed
    This graphics card has a higher base GPU clock than the average graphics card (2017 MHz vs 1920 MHz). The average graphics card has a base GPU clock of 1920 MHz.
    2,017 MHz vs 1,920 MHz
  • 2.05x higher compute score
    This graphics card has a higher compute score than the average graphics card (24,712 points vs 12,70.5 points). The average graphics card has a compute score of 12,070.5 points.
    24,712 points vs 12,070.5 points
  • 4x higher memory bandwidth
    This graphics card has a higher memory bandwidth than the average graphics card (1,792 GB/s vs 448 GB/s). The average graphics card has a memory bandwidth of 448 GB/s.
    1,792 GB/s vs 448 GB/s
  • 20 GB more VRAM
    This graphics card has more VRAM than the average graphics card (32 GB vs 12 GB). The average graphics card has 12 GB VRAM.
    32 GB vs 12 GB
  • 320 bit wider memory bus
    This graphics card has a wider memory bus than the average graphics card (512 bit vs 192 bit). The average graphics card has a memory bus width of 192 bit.
    512 bit vs 192 bit
  • 47.4% faster memory speed
    This graphics card has a higher effective memory speed than the average graphics card (28,000 MHz vs 19,000 MHz). The average graphics card reaches an effective memory speed of 19,000 MHz.
    28,000 MHz vs 19,000 MHz
  • 4x larger L2 cache
    This graphics card has more L2 cache than the average graphics card (96 MB vs 24 MB). The average graphics card has 24 MB L2 cache.
    96 MB vs 24 MB
  • Newer GDDR version
    This graphics card uses a newer GDDR version than the average graphics card (GDDR7 vs GDDR6).
    GDDR7 vs GDDR6
  • Newer PCIe version
    This graphics card supports a newer PCIe version than the average graphics card (5 vs 4).
    5 vs 4
  • 4.03x more transistors
    This graphics card has more transistors than the average graphics card (92,200 million vs 22,900 million). The average graphics card has 22,900 million transistors.
    92,200 million vs 22,900 million
  • Newer DisplayPort version
    This graphics card supports a newer DisplayPort version than the average graphics card (2.1b vs 1.4a).
    2.1b vs 1.4a
  • 4x higher HDMI bandwidth
    This graphics card has a higher HDMI bandwidth than the average graphics card (48 Gbps vs 12 Gbps). The average graphics card has HDMI bandwidth of 12 Gbps.
    48 Gbps vs 12 Gbps
  • Newer HDMI version
    This graphics card supports a newer HDMI version than the average graphics card (2.1b vs 2.1a).
    2.1b vs 2.1a
  • 3.7x higher DisplayPort bandwidth
    This graphics card has a higher DisplayPort bandwidth than the average graphics card (80 Gbps vs 21.6 Gbps). The average graphics card has DisplayPort bandwidth of 21.6 Gbps.
    80 Gbps vs 21.6 Gbps
  • 0.5 slot/s slimmer design
    This graphics card occupies fewer slots than the average graphics card (2 slot/s vs 2.5 slot/s). The average graphics card occupies 2.5 slot/s.
    2 slot/s vs 2.5 slot/s
  • 1 fewer fans
    This graphics card has fewer fans than the average graphics card (2 vs 3).
    2 vs 3
  • 2.67x higher TDP
    This graphics card has a higher TDP than the average graphics card (575 W vs 215 W). The average graphics card has a TDP of 215 W.
    575 W vs 215 W
  • No dual BIOS
    This graphics card does not include dual BIOS, the average graphics card does.
  • 3.9% lower boost clock speed
    This graphics card has a lower boost GPU clock than the average graphics card (2,407 MHz vs 2,505 MHz). The average graphics card has a boost GPU clock of 2,505 MHz.
    2,407 MHz vs 2,505 MHz
  • 53.8% higher PSU requirement
    This graphics card has a higher PSU requirement than the average graphics card (1,000 W vs 650 W). The average graphics card has a PSU requirement of 650 W.
    1,000 W vs 650 W
  • 2.73x higher board power limit
    This graphics card has a higher board power limit than the average graphics card (600 W vs 220 W). The average graphics card has a board power limit of 220 W.
    600 W vs 220 W
  • 3.56x more expensive
    NVIDIA GeForce RTX 5090 is more expensive than the average graphics card (£1,850 vs £520).
    £1,850 vs £520
  • 12 mm higher card profile
    This graphics card is taller than the average graphics card (137 mm vs 125 mm). The average graphics card has a height of 125 mm.
    137 mm vs 125 mm
  • 3.56x more expensive
    NVIDIA GeForce RTX 5090 is more expensive than the average graphics card (£1,850 vs £520).
    £1,850 vs £520
  • 14.6% worse value for money
    NVIDIA GeForce RTX 5090 has worse value for money than the average graphics card (6.47 vs 7.4).
    6.47 vs 7.41
  • 3.9% lower boost clock speed
    This graphics card has a lower boost GPU clock than the average graphics card (2,407 MHz vs 2,505 MHz). The average graphics card has a boost GPU clock of 2,505 MHz.
    2,407 MHz vs 2,505 MHz
  • 2.55x larger GPU die
    This graphics card has a higher GPU die size than the average graphics card (750 mm² vs 294 mm²). The average graphics card has a GPU die size of 294 mm².
    750 mm² vs 294 mm²
  • 2 fewer monitors per output type
    This graphics card supports fewer monitors per output type than the average graphics card (1 vs 3). The average graphics card supports 3 monitors per output type.
    1 vs 3
  • 1 fewer fans
    This graphics card has fewer fans than the average graphics card (2 vs 3).
    2 vs 3
  • 2.67x higher TDP
    This graphics card has a higher TDP than the average graphics card (575 W vs 215 W). The average graphics card has a TDP of 215 W.
    575 W vs 215 W
  • No dual BIOS
    This graphics card does not include dual BIOS, the average graphics card does.
  • 53.8% higher PSU requirement
    This graphics card has a higher PSU requirement than the average graphics card (1,000 W vs 650 W). The average graphics card has a PSU requirement of 650 W.
    1,000 W vs 650 W
  • 2.73x higher board power limit
    This graphics card has a higher board power limit than the average graphics card (600 W vs 220 W). The average graphics card has a board power limit of 220 W.
    600 W vs 220 W
  • 14.6 dB noisier under load
    This graphics card has a higher load noise level than the average graphics card (50 dB vs 35.4 dB). The average graphics card has a load noise level of 35.4 dB.
    50 dB vs 35.4 dB
  • 36 dB noisier at idle
    This graphics card has a higher idle noise level than the average graphics card (36 dB vs 0 dB). The average graphics card has an idle noise level of 0 dB.
    36 dB vs 0 dB
  • 2.5x higher idle power draw
    This graphics card has a higher idle power draw than the average graphics card (30 W vs 12 W). The average graphics card has an idle power draw of 12 W.
    30 W vs 12 W
  • 2.61x higher peak power draw
    This graphics card has a higher peak power draw than the average graphics card (575 W vs 220 W). The average graphics card has a peak power draw of 220 W.
    575 W vs 220 W
  • 13 °C higher idle temperature
    This graphics card has a higher idle temperature than the average graphics card (50 °C vs 37 °C). The average graphics card has an idle temperature of 37 °C.
    50 °C vs 37 °C
  • 6 °C higher load temperature
    This graphics card has a higher load temperature than the average graphics card (73 °C vs 67 °C). The average graphics card has a load temperature of 67 °C.
    73 °C vs 67 °C
  • 12 mm higher card profile
    This graphics card is taller than the average graphics card (137 mm vs 125 mm). The average graphics card has a height of 125 mm.
    137 mm vs 125 mm
  • 21.8 mm longer card length
    This graphics card is longer than the average graphics card (304 mm vs 282.2 mm). The average graphics card has a length of 282.2 mm.
    304 mm vs 282.2 mm
  • 97.2% heavier
    This graphics card is heavier than the average graphics card (2000 g vs 1,014 g). The average graphics card weighs 1,014 g.
    2,000 g vs 1,014 g

Video reviews

Hands-on testing, unboxing, and real-world video evaluations

7.0
iVadim
≈ estimated
18/12/2025
Hands-on review · thermals, noise and coil whine
Verdict
NVIDIA GeForce RTX 5090 is a mixed bag with excellent design and cooling but coil whine is hard to live with.
8.0
Gamers Nexus
≈ estimated
23/01/2025
Full review · benchmarks, thermals and power tests
Verdict
NVIDIA GeForce RTX 5090 offers strong 4K performance and an excellent cooler, but it is very power hungry and raises connector concerns.
7.0
Linus Tech Tips
≈ estimated
23/01/2025
Hands-on review · benchmarks, cooling and CPU temperature tests
Verdict
NVIDIA GeForce RTX 5090 is great for professionals and developers, but poor value for gamers given its price and power draw.
7.0
IGN
≈ estimated
23/01/2025
Full review · gaming and benchmark tests
Verdict
NVIDIA GeForce RTX 5090 is the fastest consumer card, but the generational gain is small outside DLSS 4-supported games.
8.0
Digital Foundry
≈ estimated
23/01/2025
Hands-on review · benchmarks and DLSS 4 analysis
Verdict
NVIDIA GeForce RTX 5090 is a big generational step and a standout card, though costly and less compelling for RTX 4090 owners.
7.0
Techtesters
≈ estimated
23/01/2025
Hands-on review · 45 games at three resolutions, thermals and power
Verdict
NVIDIA GeForce RTX 5090 is the fastest GPU you can buy, though price, power draw and coil whine count against it.
9.0
OverclockersClub
≈ estimated
23/01/2025
Hands-on review · benchmarks, thermals and AI tests
Verdict
NVIDIA GeForce RTX 5090 does its job, a notch above the RTX 4090, though the price is debatable.
9.0
Mrwhosetheboss
≈ estimated
23/01/2025
Hands-on review · Cyberpunk testing and frame generation latency
Verdict
NVIDIA GeForce RTX 5090 is easily the best-looking gaming experience, though AI frame artifacts and NVIDIA's marketing both still need work.
7.0
JayzTwoCents
≈ estimated
23/01/2025
Full review · performance, power and temperature tests
Verdict
NVIDIA GeForce RTX 5090 is excellent for professionals, but poor value for most gamers compared with the RTX 5080.
6.0
Hardware Unboxed
≈ estimated
23/01/2025
Benchmark review · gaming performance and value analysis
Verdict
NVIDIA GeForce RTX 5090 is a very impressive performer, but it is disappointing value at this price.
Hands-on video, no verdict
Not graded

Expert reviews

In-depth analysis and verdict from leading tech publications

8.0
au.pcmag.com
✓ explicit
30/01/2025
Lab test · Gaming and rendering
Verdict
The Nvidia GeForce RTX 5090 is the fastest card in this test, especially for Blender and ray-traced gaming, but it draws a lot of power.
8.0
tomshardware.com
✓ explicit
23/01/2025
Lab test · Rasterization gaming
Verdict
The GeForce RTX 5090 is about 25% faster than the RTX 4090 at 4K ultra, but early drivers leave some games slower at lower resolutions.
7.0
tomsguide.com
≈ estimated
23/01/2025
Benchmark test · DLSS 4 preview
Verdict
The GeForce RTX 5090 is about 20 to 25% faster than the RTX 4090 without DLSS, so owners of RTX 40 cards can likely skip it.
8.0
pugetsystems.com
≈ estimated
21/02/2025
Lab test · AI workloads
Verdict
The GeForce RTX 5090 is the fastest card for running AI language models locally, with 29% quicker text generation than the RTX 4090, though early drivers mar some tests.
8.0
nikolasent.github.io
≈ estimated
17/02/2025
Benchmark test · Computer vision
Verdict
The GeForce RTX 5090 is about 44% faster than the RTX 4090 at computer vision tasks and 72% faster on language models, while drawing about 28% more power.
8.0
tomshardware.com
✓ explicit
23/01/2025
Lab review · Founders Edition
Verdict
The GeForce RTX 5090 is the fastest GPU tested, but its 575W power draw, hot-running Founders Edition, and early driver issues hold it back.
8.0
gamersnexus.net
≈ estimated
24/01/2025
Lab review · Thermals and noise
Verdict
The GeForce RTX 5090 Founders Edition keeps its GPU cool in a two-slot body, but its memory runs warm and needs case airflow.
8.0
kitguru.net
✓ explicit
23/01/2025
Benchmark review · Ray tracing
Verdict
The GeForce RTX 5090 packs 32GB of GDDR7 and 575W of power, and this review tests whether Nvidia's claim of up to double the RTX 4090's speed holds.
8.0
pcmag.com
✓ explicit
29/01/2025
Lab review · Creative app tests
Verdict
The GeForce RTX 5090 leads most tests here, pulling far ahead in Blender rendering, and AMD has no consumer card that competes with it.
7.0
pugetsystems.com
≈ estimated
23/01/2025
Lab test · Creative apps
Verdict
The GeForce RTX 5090 is the strongest card here for rendering in V-Ray and Blender, though its lead in Premiere Pro editing is small.
9.0
pcper.com
≈ estimated
23/01/2025
Benchmark review · Ultrawide resolution
Verdict
The GeForce RTX 5090 runs about 20 to 40% faster than the RTX 4090 at 3440x1440 without DLSS, though frame smoothness with 4x frame generation still disappoints.
8.0
storagereview.com
≈ estimated
23/01/2025
Benchmark review · AI and gaming
Verdict
The NVIDIA GeForce RTX 5090 is a powerful but smaller flagship with strong DLSS 4 frame generation gains, though it draws much more power than the RTX 4090.
T
techspot.com
23/01/2025
✓ explicit
8.0
Summary of the review
Nvidia GeForce RTX 5090 Review | TechSpot
U
uk.pcmag.com
29/01/2025
✓ explicit
8.0
Summary of the review
Nvidia GeForce RTX 5090 Founders Edition - Review 2025 | PCMag
T
techradar.com
23/01/2025
✓ explicit
8.5
Summary of the review
Nvidia GeForce RTX 5090 review: the supercar of graphics cards | TechRadar
forbes.com
23/01/2025
Not graded yet
mmorpg.com
23/01/2025
Not graded yet
forbes.com
23/01/2025
Not graded yet
10
lesnumeriques.com
✓ explicit
23/01/2025
Hands-on test · AI frame generation
Verdict
The Nvidia GeForce RTX 5090 is a strong 4K card with impressive DLSS 4 frame generation, but it draws a lot of power and needs a very well-ventilated case.
8.0
phonandroid.com
≈ estimated
23/01/2025
Lab test · Rendering and gaming
Verdict
The Nvidia GeForce RTX 5090 Founders Edition is a big jump for 3D rendering, but it draws up to 550 W and whines at high frame rates.

Third-party reviews

What customers like about NVIDIA GeForce RTX 5090

  • Unmatched 4K gaming performance, often 30% faster than the RTX 4090
  • Massive 32GB of GDDR7 memory suitable for heavy AI and creative workloads
  • Relatively compact two-slot Founders Edition design compared to previous triple-slot cards
  • Excellent thermal management, maintaining surprisingly low temperatures under load
  • Support for DLSS 4 and advanced ray-tracing features
  • Significant improvements in video encoding with triple ninth-gen encoders

What customers don't like about NVIDIA GeForce RTX 5090

  • Extremely high MSRP of $1,999, with reseller prices often much higher
  • Significant power requirements, with total system draw reaching over 850W in some tests
  • Heavy CPU bottlenecking at lower resolutions like 1080p and 1440p
  • Lackluster generational efficiency gains compared to the RTX 4090
  • Ongoing concerns regarding the reliability and safety of the 12VHPWR power connector
  • High idle power consumption

Verdict

The Nvidia GeForce RTX 5090 is the flagship of the Blackwell architecture, featuring 21,760 CUDA cores, a boost clock of 2.41 GHz, and a massive 32 GB of GDDR7 VRAM on a 512-bit bus, delivering approximately 1.79 TB/s of memory bandwidth. Its primary advantages include class-leading 4K gaming and ray tracing performance—showing roughly 20-50% gains over the RTX 4090—unparalleled AI horsepower with 3,352 Tensor TOPS, and a surprisingly compact two-slot Founders Edition design. However, it faces significant drawbacks, such as an eye-watering $1,999 MSRP (often higher in retail), a substantial 575W power requirement that may necessitate a 1000W+ PSU, and frequent CPU bottlenecks at resolutions lower than 4K.

Compare NVIDIA GeForce RTX 5090 with other products in the Graphics cards category

1Product to compare no. 1
VS
2Product to compare no. 2
VS
3Product to compare no. 3 (optional)

Graphic comparison of NVIDIA GeForce RTX 5090 and other products in the Graphics cards category

No results found