CORSAIR RM850e (2025) ATX 3.1 PCIe 5.1 Ready 850W Power Supply - Black
The RM850e combines 850 W rated output, compact modular cabling and modern GPU connectivity. Its measured efficiency supports the Gold rating, but full-load noise and weaker 3.3V transient response mean it suits buyers who accept those compromises rather than expecting premium performance from every rail.
Updated 2026-10-08
We may earn a commission from qualifying purchases.
Prices and sellers
No price information yet
Conditions and limitations
Rated output, specified excursions and test conditions are documented; no complete-PC guarantee is established.
Compatibility
Connector inventory was reviewed; exact motherboard and GPU pair tests are not established.
Where sources differ
Certification averages and load-specific results differ in conditions. BuildCores and Scan also show inconsistent certification labels; review results are retained.
Reported issues and fixes
The review identifies weaker 3.3V transient response; affected batches, confirmed fixes and failure prevalence remain unknown.
Care and support
Seven-year warranty is recorded; regional eligibility, spare availability and a support end date are not established.
Ownership costs
Efficiency measurements inform energy use, but workload and electricity inputs are absent; no cost total is calculated.
Who it suits
Article and analysis explain capacity, cabling, noise and warranty tradeoffs; complete-system compatibility remains untested.
Measured performance
| Conditions | Value | Evidence |
|---|---|---|
| Cybenetics efficiency result reported by Hardware Busters · 115 V input. Reported certification average; load points and sample count are not specified in this table. | 88.568 % | [1] |
| KitGuru programmable DC-load test · 240 V input at 450 W DC load; the stated general test environment was 35c. Sample count was not stated. | 93.2 % | [9] |
Efficiency
88.568 %Cybenetics efficiency result reported by Hardware Busters · 115 V input. Reported certification average; load points and sample count are not specified in this table.
Source [1]Efficiency
93.2 %KitGuru programmable DC-load test · 240 V input at 450 W DC load; the stated general test environment was 35c. Sample count was not stated.
Source [9]Noise level
39.1 dBAKitGuru digital sound-level measurement · 850 W DC load, meter one meter from the PSU in the acoustics room with no other fans running. Sample count was not stated.
Source [9]The review
Corsair RM850e: power, efficiency, connectors, and warranty
Rated output and platform
The Corsair RM850e covered here is the 2025 ATX 3.1 generation, associated with the CP-9020296 model family. BuildCores lists CP-9020296-NA alongside regional variants; Scan identifies its own listing as CP-9020296-UK. The assigned US product identity stays separate from that UK listing. This distinction matters when checking what will arrive in the box: a familiar RM850e name alone is not sufficient evidence that an older generation or another regional package has the same cables. The recorded manual gives total output as 850 W, while the contemporary reviews describe ATX 3.1 support and modern GPU power connectivity. Together, these establish the supply's intended role without treating a regional catalogue as a substitute for the assigned variant. KitGuru's feature list describes support for GPU transient excursions up to twice nominal power. That is a statement about the specified transient capability, not an independent measurement of continuously available power beyond the 850 W rating. A buyer should therefore keep the rated capacity and the excursion claim as separate considerations. The practical implication is to evaluate the whole planned system rather than choosing from the graphics connector alone. Having the right connection does not establish that every possible combination of processor, graphics card and operating settings is suitable. The reviews also help set expectations within Corsair's range. KitGuru awarded this model 8/10, but an editorial score cannot be stored or interpreted as an electrical benchmark. Hardware Busters identifies a weaker 3.3V transient response, so ATX 3.1 support should not be read as proof that every rail performs equally well. For a builder comparing supplies, the useful question is whether the documented connections and the tested compromises match the intended system. This research supplies those distinctions; it does not certify an untested complete PC. [7], [2], [2], [4], [5], [6]
Efficiency and noise certification
Hardware Busters reports Cybenetics Gold efficiency and LAMBDA-A noise certification for the RM850e. Its specification table records a 115 V efficiency result of 88.568%. This is a reported certification result, not a promise that the supply operates at exactly that efficiency at every load. The inherited feature description saying up to 90% efficiency is likewise a general feature claim. Keeping the type of evidence visible prevents a marketing summary, a certification average and a particular bench result from being treated as interchangeable numbers. KitGuru provides a more specific result: its 240 V efficiency table records 93.2% at a 450 W DC load. The review describes a programmable-load test setup and a general environment of 35c. This result answers a narrower question than the 115 V figure because the input voltage and load point differ. It is useful evidence of the tested unit's performance at that operating point, but it should not be presented as a contradiction of the certification result. Nor does it establish a household electricity saving without a workload profile and the other inputs needed for that calculation. Those assumptions are not supplied here, so no ownership-cost total is inferred. Noise requires the same care. Hardware Busters explains the LAMBDA-A rating as an average acoustic range of 20–25 dBA. KitGuru, however, measured 39.1 dBA at an 850 W load. Its acoustic procedure placed the meter one meter from the PSU with no other fans running in the room. The two descriptions concern different test conditions and summaries, rather than a universal noise level. For a buyer, the important implication is that an average quietness certification does not guarantee equally quiet operation when the supply is working at full output. A quiet workstation decision should consider the intended load, not simply the noise badge. Neither figure measures the finished computer with its other cooling components. [7], [2], [1], [1], [9]
Cable connectivity
The review's cable inventory includes two EPS cables, three conventional PCIe 6+2 connectors and one high-power GPU connector described as 600 W capable. These are useful planning details, but connector count and separate cable count are different quantities. A builder should compare the intended motherboard and graphics card requirements with the actual supplied harnesses, rather than assuming that every connector listed represents an independent cable. The contemporary feature list also identifies the native 12V-2x6 connection. The inherited review inventory uses a different pin shorthand, so the model generation and named connector standard are more useful purchasing checks than that shorthand by itself. Fully modular cabling means only the cables needed by the build have to be connected, as KitGuru's feature description explains. For a relatively simple system, that gives the builder a clearer way to plan routing without installing every supplied harness. It does not establish that a cable from another supply is interchangeable, and this research makes no such compatibility claim. The safe purchasing decision is to confirm the package and the manufacturer's applicable cable guidance for the particular model before relying on an existing cable collection. The documented 140 mm casing length is another practical consideration. It supports describing the unit as compact, but it is not a measurement of the free space remaining in a particular case. A prospective owner still needs to assess the chosen enclosure and the room required around the cable connections. KitGuru's teardown identifies a Hong Hua 120 mm fan, which supports the cooling-construction description rather than a prediction of noise in any enclosure. These details make fit assessment more concrete: check the casing, harnesses and system requirements together. Finally, the UK retailer listing can corroborate the generation's general specifications, while the CP-9020296-NA mapping keeps regional identity explicit. UK package or service details are not silently transferred to the US product. [1], [2], [3], [5], [6], [1]
Warranty tradeoff
Hardware Busters records a seven-year warranty, and KitGuru identifies warranty length as a compromise relative to some premium models it reviewed. The duration is meaningful for a buyer intending to keep a supply through several builds, but it is not a measured service life. It also does not by itself establish the handling of every future warranty claim. Regional eligibility, exclusions and service procedures should be checked against the applicable terms. This research preserves the recorded duration while leaving those unsupported details open, rather than treating a UK retailer's service description as US policy. The independent electrical caveat is more specific than a general statement that this is a lower-tier product. Hardware Busters found mediocre transient response on the 3.3V rail and wanted protection thresholds adjusted. That observation belongs to the reviewed sample and test context. It does not prove that every delivered unit is defective, and no confirmed hardware fix or affected serial-number range is established here. The useful implication is that a buyer prioritizing the strongest performance across all rails should examine detailed tests instead of relying only on output capacity and certification. Those who primarily need the documented connections should still understand what the review identified as a compromise. Taken together, the evidence supports a conditional fit for a modern build needing 850 W rated capacity, the documented GPU connections and a compact casing. Gold certification is supported, while actual full-load acoustic testing puts a boundary on expectations of quietness. The seven-year warranty and the minor-rail caveat belong in the decision alongside those strengths. KitGuru's 8/10 verdict provides editorial context, but the measurable results and their conditions are the more useful basis for a technical comparison. Before buying, confirm the exact regional model, the required cable arrangement, enclosure clearance and the whole system's power needs. This conclusion stays within the available tests and documentation; it does not promise compatibility, silence or lifetime reliability for a system that was never tested. [7], [2], [1], [2], [9], [4]
Buyer questions
Before you buy
What efficiency certification is reported?
Hardware Busters reports Cybenetics Gold efficiency certification, with a 115 V result of 88.568%. That is not efficiency at every load. [1]
Compatibility
What is the RM850e's rated total power?
Corsair's manual lists total power at 850 W. [7]
Does it support modern GPU transients?
The ATX 3.1 design is described as supporting substantial GPU transient excursions. [2]
Setup
What connector types are included?
The cable set includes EPS cables, conventional PCIe connectors, and a high-power PCIe connector. [1]
Product images
Facts and specifications
Manufacturer manual
| Specification | Value | Source |
|---|---|---|
| Power | 850 W | [7] |
Retailer listing
Stated by reviewers
| Specification | Value | Source |
|---|---|---|
| Depth | 140 mm | [2] |
| Fan Size | 120 mm | [3] |
| Limitation | Hardware Busters found mediocre 3.3V transient response and wanted protection thresholds adjusted. | [4] |
| Modularity | Fully modular | [2] |
Measured by a lab
KnowSpec analysis
Modern connections with measured limits
The RM850e combines 850 W rated output, compact modular cabling and modern GPU connectivity. Its measured efficiency supports the Gold rating, but full-load noise and weaker 3.3V transient response mean it suits buyers who accept those compromises rather than expecting premium performance from every rail.
Method: Synthesis of independent tests and documented features; buyer implication is editorial inference.Visible information gaps
Missing information is shown rather than guessed.
- Setup Requirementrights restricted
Sources and method
- Corsair RM850e ATX v3.1 PSU ReviewHardware Busters · retrieved 2026-10-08Lab review
- Corsair RM850e ATX 3.1 (2025) PSU ReviewKitGuru · retrieved 2026-10-08Editorial review
- RM850e review: component analysisKitGuru · retrieved 2026-10-08Editorial review
- RM850e review: EpilogueHardware Busters · retrieved 2026-10-08Lab review
- RM850e (2025) CP-9020296-UKScan · retrieved 2026-10-08Retailer listing
- Corsair RM850e (2025)BuildCores · retrieved 2026-10-08Other source
- RMe Series Fully-Modular ATX Power Supply ManualCorsair · retrieved 2026-09-27ManualPDF
- Corsair RM850e ATX 3.1 (2025) PSU ReviewKitGuru · retrieved 2026-09-27Editorial review
- RM850e review: testingKitGuru · retrieved 2026-10-08Lab review
- Corsair RM850e ATX v3.1 PSU ReviewHardware Busters · retrieved 2026-09-27Lab review