Dylan and the FBC might be going through it, but Remedy is flexing all its finest talents: sculpting uncanny hellscapes, placing tactical redactions in bureaucratic lore collectibles, and capturing endearingly human performances in circumstances beyond human comprehension. Unfortunately, melee combat isn't on that list of strong suits—and Resonant's open world expects you to do a lot of it.

New weird

Remedy's environmental artists and lighting tech specialists are at the height of their powers in Resonant: The stark, brutalist surrealism that saturated the Oldest House in the first game has boiled out of its breached walls and onto New York's streets, reconstructing blocks of brownstones and downtown shopping districts into a sprawling, dreamlike mazework of impossible architecture and alternative physics.

Roadways have abruptly become cliff faces, with pavement—and the buses parked on it—bent skyward at perpendicular angles. Rows of pixel-stretched structures bleed into chasms where chunks of cityscape have simply fallen away.

Whole neighborhoods spiral upwards, outwards, inwards. Some are swallowed in the telltale oilslick haze of Hiss corruption. Others have been reconfigured according to the alien logic of the Patterning—a newer, stranger force whose spreading influence sorts apartments, taxis, streetsigns, pigeons, and unevacuated denizens into inscrutable designs of sprawling radial symmetry.

It's arresting, at turns breathtakingly incomprehensible and queasily familiar. Nestled into the mutated Manhattan's warped streets are pockets of untouched city life: abandoned cafes and laundromats and bespoke art galleries, which only heighten the sense of eeriness through their proximity to reality's active unspooling.

It's gorgeous, a sequel that's begging for an inevitable photo mode update. And it's a pleasure to move through, with Dylan's accumulating collection of mobility superpowers—floating, air-dashing, double-jumping, and eventually, freely reorienting gravity—allowing for truly impressive feats of traversal design.

I just wish I could say the same for all the hours of fighting I had to do.

Break in the pattern

I wanted to enjoy the concept of Resonant's combat. While Jesse wielded a manifold pseudogun in the OG Control, Dylan employs a different sort of shape-shifting weapon technique courtesy of the object his sister impaled in his sternum. It's called the Aberrant, and in his hands, it can assume and discard the shapes of different melee weapons on demand: Depending on your preference, it can be a hatchet; a scythe, a pair of short swords, a hammer, a drill, an orbiting, levitating sawblade, and more.

Playing a guy who can throw rocks with his brain and has a stick that can be infinite swords should be cool. In practice, it's… aggressively fine.

While eventual unlocks add necessary depth and variety, Resonant's opening hours suffer from the cardinal sin of its combat design: It's an action game where basic attacks don't have the juice. On its own, pushing the attack button won't achieve much except lightly grazing the heavy enemies that need to be killed to finish each fight. Instead, Dylan has to pile on temporary damage bonuses—granted most easily by staggering and executing lesser enemies—to make swinging his magical mystery club worthwhile.

Resonant's handful of boss fights are more engaging threats to navigate (though the sheer heft of their health bars still feels like a slog). But the endless rounds of generic creep culling that make up the bulk of playtime don't provide particularly interesting problems to solve. Instead, they're the same predicament with a binary outcome: Either you've stacked enough damage buffs that hitting a heavy enemy will accomplish something, or you can spend your time whittling away at its health bar until you get a chance to set yourself up properly.

To be fair, sometimes a guy will have a forcefield on him. This provides the benefit of occasionally making things more annoying.

Still, if all the pieces of Dylan's combat kit are moving with the right rhythm, it can deliver sparks of satisfaction. When there's nothing to trip up my flow of staggering fodder, buffing myself with their execution, and tearing into heavy enemies with damage-boosted obsidian daggers and stone-clad fists, Dylan's capacity for carnage makes an enjoyable spectacle as he launches from one short-lived target to the next.

Something inevitably will trip up our boy, however: My last ounce of power—the resource that fuels my combat abilities like rock telekinesis or toggleable fire damage—will be wasted when inconsistent auto-targeting flings my boulder at the wrong enemy. I'll earn an expendable damage buff from a perfect dodge just for Dylan to lunge past the Hiss horror I intended to merc and spend it whiffing at empty air.

I'll be staggered enough times by the sheer volume of Mold monster projectiles that I'll miss my peak damage Goldilocks window. An enemy will teleport away just as I'm winding up a triple-buffed finishing combo. Or, after watching the same enemy's execution animation a few dozen too many times, it'll just feel like another few seconds separating me from the parts of Control I'm actually interested in seeing.

That said, I'm still willing to award points for the fact that the Board—the FBC's unknowable astral plane patrons—celebrate every one of Dylan's executions with an exultant, interdimensional honk.

Altered items

Even without those irritations, the spectacle of Dylan's repetitive executions can't sustain Resonant's 30+ hour playtime. Its combat isn't compelling enough to seek out—but it expects that I'm eager to do so. As I beelined towards the story leads and sidequests that most caught my interest, I'd enter areas where the enemies had scaled past my current level—a nebulous figure calculated from my current number of weapon upgrades, spent talent points, and collectible health and power increases.

An underleveled Dylan makes Resonant's damage-sponginess even more pronounced, forcing me to circle back to combat encounters and world activities I'd intentionally noped out of just to get the Faden family's sweet, soft killing machine back up to par. But even that progression scheme feels confused. Keeping pace with enemy scaling provided an excess of weapon upgrade materials and talent points—resources I needed to spend to level up, but had regularly run out of interesting options on which to spend them.

I quickly settled on a comfortable and capable set of weapon forms; once those were maxed out, any additional resources I gathered were spent on weapon shapes I never needed. Meanwhile, the glut of talent points I earned meant I was regularly choosing talents for abilities I didn't even have access to. By the time I unlocked the game's final combat ability, I'd already been upgrading it for hours because I had few other options left on the tree.

As progression lost its luster, the late game's inflating enemy health pools and damage resistances stifled whatever welcome Resonant's combat hadn't already worn out. Mercifully, its granular difficulty adjustment settings helped soothe the chafing: Once I'd seen all the moving pieces in Dylan's combat arsenal, I simply juiced his outgoing damage and stagger buildup to speed combat along in the story's last few quests. I recommend anyone else do the same—ideally, earlier than I did.

While carving up Hiss by the hundreds didn't hold my interest, Resonant's world did. Not all of its open world attractions are winners; I'm still not sure I understand one particular side activity I kept completing by aimlessly hurling boulders at rampant bits of junk. But I kept returning to Resonant for the privilege of existing within it.

It's not just because the world looks good—never before has the texture of concrete been so celebrated—but it's dense with intention that makes poking at its corners feel rewarding.

From the placement of blacked-out bureau reports to the deliberate layering of gravity-bending traversal routes, the pockets of Resonant's Manhattan that stage its puzzles, side quests, and collectibles feel hand-crafted and considered where so many open world spaces feel perfunctory. That exploration might cash out in another chest of materials or another wave of Hiss guys to delete, but as long as I got to soak in another FBC memo, bespoke apartment interior, or ominous arrangement of questionably mundane objects, I got the prize I wanted.

Human resources

Story-wise, the sequel can't fascinate in quite the same way Control could before it was a known quantity. But the embattled, functionalist footholds the FBC are fighting and failing to maintain throughout the city have an altogether different effect. Seeing the organization forced to abandon the redaction-heavy deniability of its paranatural purview to operate in the open and on the backfoot makes its mission of managing interdimensionally cursed radios and hellgate slide projectors feel impressively real.

That's thanks in large part to the talent and direction of Resonant's voice cast. From its prime players to its nameless, proximity-barking NPCs, its performances feel so thoroughly, believably exhausted: Since the first game, half of the FBC has been isolated in lockdown within the Oldest House for seven years, only reconnecting with their bewildered and directionless fellows in the outside world when this new, unknowable apocalypse breaches containment.

And I can feel it. There's a natural, organic cadence to the cast's line deliveries, making Resonant's writing read as genuine—whether its characters are overwhelmed to the point of breakdown by the scale of the crisis, entering a reverie over the metaphysical implications of a murderous stereo, or seething at the presence of the man who killed scores of their comrades.

Speaking of: Let's talk about my sweet, slaughter-stained son, Dylan. Having endured repeated tragedy, isolation, unethical experimentation, and hostile interplanar possession for most of his life, he's effectively a good-hearted 10 year old in the body of a 30-something year old man who can manipulate profound paranatural forces. I might have preferred another outing in Jesse's shoes, but I warmed up to the younger Faden sibling more than I expected—aided by his growing, stumbling rapport with his impromptu FBC handler, Special Agent Zoe De Vera.

Thanks to his compound of sheltered naivety and his regret over being an implement of attempted omnicide, Dylan's the best open world protagonist the FBC could hope to have fall into its lap: Even if he was worldly enough to know when not to offer his help, he's too guilt-ridden to withhold it. And as the game delves deeper into his past as the subject of the FBC's experimentation under all-timer bastard Dr. Casper Darling, he's an excellent lens for exploring the fallout of an organization that knows just enough to be dangerous to all of reality.

If only that fallout was more fun to fight.

I've got to the point where I grimace when I see a listing for a gaming PC. The memory crisis has so severely gutted pricing and availability that I've found my eyes glazing over PCs whenever they're on the screen in front of me.

One glance at our roundup of the best gaming PC deals shows that even getting a rig under $1,000 with a discrete graphics card is a bit of a struggle, and you are making significant compromises to get the price that… low. Okay, yes, it doesn't feel low, but that's the new normal. All praise our new AI overlords.

And iBuyPower routinely features among the most budget rigs. That's because, be it smart compromises, the ability to buy in bulk, or a prayer to some sort of hardware god, its rigs are somewhat reasonable for prebuilds.

The RDY Scale B06 I've had across my desk shoves a budget Intel CPU and budget Nvidia GPU into its surprisingly large case, alongside a pretty small SSD and single-channel memory.

Features

RDY is iBuyPower's preconfigured range, which means they are ready to ship pretty much the moment you hit 'buy'. As such, the RTX 5060 / Intel Core Ultra 5 225F model I've got my hands on is the only version of the B06 you can currently buy.

For its relatively budget-friendly price, it makes some compromises, the biggest of which you will find in its memory and storage.

The memory is DDR5, but with a CAS latency of 40 and overall speed of 5200 MT/s, it's not super fast. There's also only one stick, so you are taking a hit to performance over dual-channel memory of the same capacity. How much depends on the game or workload, but there will be a hit.

Should you have the funds and the patience, you could technically track down a second module matching this same stick and bump your capacity up to 32 GB, but that's a hard sell, and was even before the memory crisis. Most of the time, you are better off selling a single stick and buying two (at a loss) if you're looking to upgrade, as it can be tough to track down the exact stick you need and even that doesn't guarantee compatibility.

There's just 1 TB of storage in this machine. That is pretty much the minimum I would want in a gaming rig right now, especially when you consider the actual storage available to you will likely be 50-100 GB lower than 1 TB.

The 1 TB in my rig translates to 952 GB. That doesn't include browsers and any other Windows apps. Then, you might have Steam, Epic Games, likely the Nvidia app, plus any drivers, and any other media apps like Spotify. By only downloading the games I used for the benchmarks below, I was left with around 300 GB on the drive.

Now, we would have received twice the memory capacity for the price if it weren't for the memory crisis, but you have to expect to pay more for gaming PCs right now, and those compromises make some sense from iBuyPower.

Performance

This rig is a sign of how bad the memory crisis has gotten. Late last year, we checked out the Acer Predator Orion 3000, which was just $20 more, came with a full-fat RTX 5070, a better Intel chip for multithreaded workloads, and even came in a cute compact case.

Because of that higher-tier GPU, the Orion outpaces this iBuyPower rig in every gaming test I conducted. Some by just a few frames, and some more convincingly so. Acer did make compromises here with PCIe 3.0 storage speed and 16 GB DDR4 memory, but it's still all in all a better rig, and a much more compact one. The caveat, though, is that while you can find RTX 5070 PCs at this price on occasion, they are becoming ever more rare, and this SKU of the Orion 3000 is nowhere to be seen these days.

1440p gaming performance

This chart shows average and 1% low frame rate results in various tested games at 1440p resolution

Avatar: Frontiers of Pandora (1440p Ultra)Cyberpunk 2077 (1440p RT Medium)Cyberpunk 2077 (upscaling) (1440p RT Medium)Baldur's Gate 3 (1440p Ultra)Black Myth: Wukong (1440p Very High)Black Myth: Wukong (upscaling) (1440p Very High)Metro Exodus: Enhanced Edition (1440p Ultra)

Avg FPS

1% Low FPS

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

42

33

Acer Predator Orion 3000 I RTX 5070 I 14700F

88

43

AtomMan G1 Pro | RTX 5060 | 8945HX

44

38

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

68

27

iBuyPower RDY Scale B06

36

24

0255075100

Avatar: Frontiers of Pandora (1440p Ultra) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	42 Avg FPS, 33 1% Low FPS
Acer Predator Orion 3000 I RTX 5070 I 14700F	88 Avg FPS, 43 1% Low FPS
AtomMan G1 Pro | RTX 5060 | 8945HX	44 Avg FPS, 38 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	68 Avg FPS, 27 1% Low FPS
iBuyPower RDY Scale B06	36 Avg FPS, 24 1% Low FPS

Avg FPS

1% Low FPS

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

32

25

Acer Predator Orion 3000 I RTX 5070 I 14700F

57

40

AtomMan G1 Pro | RTX 5060 | 8945HX

32

26

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

66

55

iBuyPower RDY Scale B06

35

28

020406080

Cyberpunk 2077 (1440p RT Medium) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	32 Avg FPS, 25 1% Low FPS
Acer Predator Orion 3000 I RTX 5070 I 14700F	57 Avg FPS, 40 1% Low FPS
AtomMan G1 Pro | RTX 5060 | 8945HX	32 Avg FPS, 26 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	66 Avg FPS, 55 1% Low FPS
iBuyPower RDY Scale B06	35 Avg FPS, 28 1% Low FPS

Avg FPS

1% Low FPS

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

50

42

Acer Predator Orion 3000 I RTX 5070 I 14700F

139

49

AtomMan G1 Pro | RTX 5060 | 8945HX

86

69

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

184

102

iBuyPower RDY Scale B06

90

67

050100150200

Cyberpunk 2077 (upscaling) (1440p RT Medium) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	50 Avg FPS, 42 1% Low FPS
Acer Predator Orion 3000 I RTX 5070 I 14700F	139 Avg FPS, 49 1% Low FPS
AtomMan G1 Pro | RTX 5060 | 8945HX	86 Avg FPS, 69 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	184 Avg FPS, 102 1% Low FPS
iBuyPower RDY Scale B06	90 Avg FPS, 67 1% Low FPS

Avg FPS

1% Low FPS

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

90

58

Acer Predator Orion 3000 I RTX 5070 I 14700F

73

38

AtomMan G1 Pro | RTX 5060 | 8945HX

59

42

iBuyPower RDY Scale B06

71

43

0255075100

Baldur's Gate 3 (1440p Ultra) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	90 Avg FPS, 58 1% Low FPS
Acer Predator Orion 3000 I RTX 5070 I 14700F	73 Avg FPS, 38 1% Low FPS
AtomMan G1 Pro | RTX 5060 | 8945HX	59 Avg FPS, 42 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	N/A
iBuyPower RDY Scale B06	71 Avg FPS, 43 1% Low FPS

Avg FPS

1% Low FPS

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

29

24

Acer Predator Orion 3000 I RTX 5070 I 14700F

46

31

AtomMan G1 Pro | RTX 5060 | 8945HX

26

20

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

56

49

iBuyPower RDY Scale B06

30

23

015304560

Black Myth: Wukong (1440p Very High) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	29 Avg FPS, 24 1% Low FPS
Acer Predator Orion 3000 I RTX 5070 I 14700F	46 Avg FPS, 31 1% Low FPS
AtomMan G1 Pro | RTX 5060 | 8945HX	26 Avg FPS, 20 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	56 Avg FPS, 49 1% Low FPS
iBuyPower RDY Scale B06	30 Avg FPS, 23 1% Low FPS

Avg FPS

1% Low FPS

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

49

41

Acer Predator Orion 3000 I RTX 5070 I 14700F

70

48

AtomMan G1 Pro | RTX 5060 | 8945HX

37

33

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

83

51

iBuyPower RDY Scale B06

47

39

0255075100

Black Myth: Wukong (upscaling) (1440p Very High) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	49 Avg FPS, 41 1% Low FPS
Acer Predator Orion 3000 I RTX 5070 I 14700F	70 Avg FPS, 48 1% Low FPS
AtomMan G1 Pro | RTX 5060 | 8945HX	37 Avg FPS, 33 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	83 Avg FPS, 51 1% Low FPS
iBuyPower RDY Scale B06	47 Avg FPS, 39 1% Low FPS

Avg FPS

1% Low FPS

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

61

42

Acer Predator Orion 3000 I RTX 5070 I 14700F

97

27

AtomMan G1 Pro | RTX 5060 | 8945HX

52

36

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

115

42

iBuyPower RDY Scale B06

65

45

037.575112.5150

Metro Exodus: Enhanced Edition (1440p Ultra) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	61 Avg FPS, 42 1% Low FPS
Acer Predator Orion 3000 I RTX 5070 I 14700F	97 Avg FPS, 27 1% Low FPS
AtomMan G1 Pro | RTX 5060 | 8945HX	52 Avg FPS, 36 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	115 Avg FPS, 42 1% Low FPS
iBuyPower RDY Scale B06	65 Avg FPS, 45 1% Low FPS

The RTX 5060 is a solid 1080p card that becomes a decent 1440p one if you are willing to stomach a bit of frame generation or upscaling. Still, it's one of the weakest offerings of the RTX Blackwell generation.

In our tests, I spotted similar scores to RTX 4060 and RTX 4060 Ti rigs, and that's partly because this same price point would have gotten you better CPUs a few years ago to throw alongside it, but also because the generational leap between the RTX 40- and RTX 50-series isn't monumental.

The RTX 5060 / Intel Core Ultra 5 225 F combo is decidedly budget, as far as modern offerings are concerned, and you will spot that in our gaming tests. Avatar: Frontiers of Pandora at Ultra on 1440p barely cracks 30 fps, and the same is true for Cyberpunk 2077 and Black Myth: Wukong. Upscaling makes these games much more playable, and the older Metro Exodus can stay around 60 fps.

The AtomMan G1 Pro, with its low-profile RTX 5060 and AMD Ryzen 9 8945HX, outperforms it in Avatar, but the Scale B06 takes the win in everything else. The G1 Pro is kitted out with 32 GB of DDR5 5200 memory. That is double the capacity of the Scale B06, and dual-channel. That means it has not only twice the capacity but twice the available bandwidth, too. But the Intel chip helps the iBuyPower PC pull through for the win.

Notably, the 1% lows are almost the same for both rigs in Baldur's Gate 3, despite the Scale B06 easily outpacing the G1 Pro in average fps. That's likely down to the memory. Baldur's Gate 3 is quite a CPU-heavy game, and the slower, lower-capacity RAM is hindering what should be a much faster chip in the Scale B06.

The Intel CPU in the Scale B06 has fewer physical cores and a lower boost clock than the 8645HX, but higher base clock speeds and a great L1 and L2 cache. In Cinebench, I saw a slightly higher single-core index for the Intel chip, but a significantly better multithread index for the AMD one. The Atomman G1 Pro has, however, gone up significantly in price even since our review in April.

The RTX 4060 Ti ABS Cyclone Aqua we reviewed last year seems to outpace or match this 5060 rig in most tests, only firmly losing in the Black Myth: Wukong upscaling benchmark. And with the RTX 5060, you pretty much need to get used to some level of upscaling if you want a smooth experience in modern games. Upscaling technology has gotten much better over time, and so has frame generation, but it's still a pretty weak card as far as pure rasterisation is concerned.

System performance

The laptops are here tested against intensive benchmarks that mostly test the CPU, but also the GPU, for productivity tasks.

3DMark Time Spy Extreme3DMark StorageCinebench 2024Blender 4.2.0 (junkshop)7zip 24.07Handbrake UHD to 1080p conversion

CPU index

GPU index

Overall index

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

10593

6124

6555

Acer Predator Orion 3000 I RTX 5070 I 14700F

7307

10663

9975

AtomMan G1 Pro | RTX 5060 | 8945HX

9435

6380

6705

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

6031

14162

11779

iBuyPower RDY Scale B06

6051

6536

6458

03,7507,50011,25015,000

3DMark Time Spy Extreme Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	10593 CPU index, 6124 GPU index, 6555 Overall index
Acer Predator Orion 3000 I RTX 5070 I 14700F	7307 CPU index, 10663 GPU index, 9975 Overall index
AtomMan G1 Pro | RTX 5060 | 8945HX	9435 CPU index, 6380 GPU index, 6705 Overall index
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	6031 CPU index, 14162 GPU index, 11779 Overall index
iBuyPower RDY Scale B06	6051 CPU index, 6536 GPU index, 6458 Overall index

Overall index

Avg bandwidth (MB/s)

Access time (us)

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

2155

381.13

86

Acer Predator Orion 3000 I RTX 5070 I 14700F

1525

261.85

118

AtomMan G1 Pro | RTX 5060 | 8945HX

1493

259.42

122

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

2343

403.37

77

iBuyPower RDY Scale B06

2152

371.21

84

07501,5002,2503,000

3DMark Storage Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	2155 Overall index, 381.13 Avg bandwidth (MB/s), 86 Access time (us)
Acer Predator Orion 3000 I RTX 5070 I 14700F	1525 Overall index, 261.85 Avg bandwidth (MB/s), 118 Access time (us)
AtomMan G1 Pro | RTX 5060 | 8945HX	1493 Overall index, 259.42 Avg bandwidth (MB/s), 122 Access time (us)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	2343 Overall index, 403.37 Avg bandwidth (MB/s), 77 Access time (us)
iBuyPower RDY Scale B06	2152 Overall index, 371.21 Avg bandwidth (MB/s), 84 Access time (us)

Single core index

Multi thread index

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

124

1668

Acer Predator Orion 3000 I RTX 5070 I 14700F

114

1024

AtomMan G1 Pro | RTX 5060 | 8945HX

111

1323

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

110

1046

iBuyPower RDY Scale B06

119

802

05001,0001,5002,000

Cinebench 2024 Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	124 Single core index, 1668 Multi thread index
Acer Predator Orion 3000 I RTX 5070 I 14700F	114 Single core index, 1024 Multi thread index
AtomMan G1 Pro | RTX 5060 | 8945HX	111 Single core index, 1323 Multi thread index
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	110 Single core index, 1046 Multi thread index
iBuyPower RDY Scale B06	119 Single core index, 802 Multi thread index

CPU (samples)

GPU (samples)

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

130

920

Acer Predator Orion 3000 I RTX 5070 I 14700F

72

1642

AtomMan G1 Pro | RTX 5060 | 8945HX

132

953

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

83

918

iBuyPower RDY Scale B06

64

931

05001,0001,5002,000

Blender 4.2.0 (junkshop) Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	130 CPU (samples), 920 GPU (samples)
Acer Predator Orion 3000 I RTX 5070 I 14700F	72 CPU (samples), 1642 GPU (samples)
AtomMan G1 Pro | RTX 5060 | 8945HX	132 CPU (samples), 953 GPU (samples)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	83 CPU (samples), 918 GPU (samples)
iBuyPower RDY Scale B06	64 CPU (samples), 931 GPU (samples)

Compressing (GIPS)

Decompressing (GIPS)

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

164

182

Acer Predator Orion 3000 I RTX 5070 I 14700F

84

119

AtomMan G1 Pro | RTX 5060 | 8945HX

87

199

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

115

131

iBuyPower RDY Scale B06

70

78

050100150200

7zip 24.07 Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	164 Compressing (GIPS), 182 Decompressing (GIPS)
Acer Predator Orion 3000 I RTX 5070 I 14700F	84 Compressing (GIPS), 119 Decompressing (GIPS)
AtomMan G1 Pro | RTX 5060 | 8945HX	87 Compressing (GIPS), 199 Decompressing (GIPS)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	115 Compressing (GIPS), 131 Decompressing (GIPS)
iBuyPower RDY Scale B06	70 Compressing (GIPS), 78 Decompressing (GIPS)

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

88

Acer Predator Orion 3000 I RTX 5070 I 14700F

42

AtomMan G1 Pro | RTX 5060 | 8945HX

80

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

75

iBuyPower RDY Scale B06

57

0255075100

Handbrake UHD to 1080p conversion Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	88
Acer Predator Orion 3000 I RTX 5070 I 14700F	42
AtomMan G1 Pro | RTX 5060 | 8945HX	80
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	75
iBuyPower RDY Scale B06	57

Outside of gaming, the Team Group T-Force G50 SSD is decently performant and speedy, clocking in an overall index score of well over 2,000 in 3DMark's storage test. Other creator tests don't perform quite as well, with the CPU at this machine's heart being fairly slow to handle a Handbrake video conversion, and sitting well behind the likes of the Aqua and G1 Pro in Time Spy Extreme and 7zip.

Though single-core Cinebench scores are solid, multithread scores are lacking. This all means that if you're looking for a heavy-duty content creator machine, you'd best be looking elsewhere. In fairness, the single-channel memory should have already put you off, anyways.

Thermals

Real-world temperatures

temperature results recorded in different testing suites

Gaming TempsCreator tempsIdle temps

Avg CPU Temp (°C)

Max CPU Temp (°C)

Avg GPU Temp (°C)

Max GPU Temp (°C)

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

62

79

64

40

Acer Predator Orion 3000 I RTX 5070 I 14700F

63

92

65

45

AtomMan G1 Pro | RTX 5060 | 8945HX

89

91

77

67

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

63

71

60

43

iBuyPower RDY Scale B06

53

59

65

67

0255075100

Gaming Temps Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	62 Avg CPU Temp (°C), 79 Max CPU Temp (°C), 64 Avg GPU Temp (°C), 40 Max GPU Temp (°C)
Acer Predator Orion 3000 I RTX 5070 I 14700F	63 Avg CPU Temp (°C), 92 Max CPU Temp (°C), 65 Avg GPU Temp (°C), 45 Max GPU Temp (°C)
AtomMan G1 Pro | RTX 5060 | 8945HX	89 Avg CPU Temp (°C), 91 Max CPU Temp (°C), 77 Avg GPU Temp (°C), 67 Max GPU Temp (°C)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	63 Avg CPU Temp (°C), 71 Max CPU Temp (°C), 60 Avg GPU Temp (°C), 43 Max GPU Temp (°C)
iBuyPower RDY Scale B06	53 Avg CPU Temp (°C), 59 Max CPU Temp (°C), 65 Avg GPU Temp (°C), 67 Max GPU Temp (°C)

Max CPU temp (°C)

Max SSD temp (°C)

Max VRM temp (°C)

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

91

67

70

Acer Predator Orion 3000 I RTX 5070 I 14700F

69

27

AtomMan G1 Pro | RTX 5060 | 8945HX

93

58

71

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

80

46

iBuyPower RDY Scale B06

67

53

0255075100

Creator temps Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	91 Max CPU temp (°C), 67 Max VRM temp (°C), 70 Max SSD temp (°C)
Acer Predator Orion 3000 I RTX 5070 I 14700F	69 Max CPU temp (°C), 27 Max SSD temp (°C)
AtomMan G1 Pro | RTX 5060 | 8945HX	93 Max CPU temp (°C), 58 Max VRM temp (°C), 71 Max SSD temp (°C)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	80 Max CPU temp (°C), 46 Max SSD temp (°C)
iBuyPower RDY Scale B06	67 Max CPU temp (°C), 53 Max SSD temp (°C)

CPU (°C)

GPU (°C)

ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D

27

35

Acer Predator Orion 3000 I RTX 5070 I 14700F

33

22

AtomMan G1 Pro | RTX 5060 | 8945HX

51

38

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D

42

37

iBuyPower RDY Scale B06

33

43

015304560

Idle temps Data

Product	Value
ABS Cyclone Aqua | RTX 4060 Ti | 7800X3D	27 CPU (°C), 35 GPU (°C)
Acer Predator Orion 3000 I RTX 5070 I 14700F	33 CPU (°C), 22 GPU (°C)
AtomMan G1 Pro | RTX 5060 | 8945HX	51 CPU (°C), 38 GPU (°C)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | 7800X3D	42 CPU (°C), 37 GPU (°C)
iBuyPower RDY Scale B06	33 CPU (°C), 43 GPU (°C)

Perhaps owing to the large capacity of the case allowing smooth, unblocked airflow, or partly due to the large fan on the CPU, or maybe because both the CPU and GPU are fairly low wattage, this rig is positively frosty. Measuring temperatures during our range of creator applications, it is among the coolest we've tried in recent times, sitting well below the iBuyPower RDY Element 9 Pro R07. Admittedly, the latter iBuyPower machine is much more powerful, but its max CPU temps are significantly under the RTX 4060-equipped Lenovo Legion Tower 5i (Gen 8), the RTX 4060 Ti ABS Cyclone Aqua, and the RTX 5060 AtomMan G1 Pro. In fairness, the AtomMan is a very compact build, so you would expect higher thermals.

Still, the point stands that this iBuyPower handles its heat in a particularly cool fashion. Unfortunately, my model's motherboard doesn't appear to have a VRM temp sensor, so I couldn't grab those temperatures, but the CPU and SSD heats were very manageable in the creator suite.

For gaming, it's much the same story, with the GPU only maxing out at 67 °C. This is particularly impressive because it averaged 65 °C. It's a similar story with the CPU, with both average and max sitting in the 50s.

Aesthetics

The Scale B06 feels like an oddly large PC for what inside is a fairly compact build. You don't have extra coolers hogging up space here, and the Asus RTX 5060 is a fairly small card, too. That means there is tons of empty space in the case. Its dimensions make it fairly similar to the Maingear MG-1 MK.II, but at least the Maingear with its overhead cooling and large RTX 5090 fits the space a bit better. Still, I suppose that means you can throw a few Hatsune Miku figures in the RDY Scale B06 and I'd be confident they won't melt.

Other than the RGB fans and the iBuyPower logo slapped on the bottom of the front panel, the Scale B06 is fairly plain. It's just a big black case with a clear side panel. Though the RGB in the fans do give it a bit of a gamer look. My model has pretty great wiring, all funnelling through the back of the rig tidily, so it's super clean.

The first time I unboxed it, the Scale B06 was a bit of a task to get on the PC Gamer test bench table. It's not only bulky, but quite heavy at 16 kg too. You are going to need quite a bit of space (and a few muscles) to fit it into your gaming den.

One thing that surprised me when taking the panels of this PC off is that they aren't held in with screws. They are simply wedged into the side, but they require a bit of a yank to come out, which makes me feel comfortable that they won't just pop off and spread broken tempered glass all over the floor when a breeze comes through the office.

Sound

As far as sound is concerned, the B06 offers a downright pleasant experience, only ever rising to a light hum at its most intensive. That translates to around 40 decibels, which is light enough that even a relatively quiet office will drown it out. In a totally silent environment, you will definitely hear the Scale B06, but you have to accept that with pretty much any bit of hardware, especially one with so many parts.

Even when I put my ear directly against the tempered glass while the rig was working its hardest, I could only make out a soft hum from the fans.

Software

My review rig showed up with a fresh Windows install, and the only other preinstalled software was Realtek audio drivers, Intel drivers, and the Nvidia app. iBuyPower doesn't appear to load the Scale B06 with its own software, and that's a boon given the size of modern PC games and the capacity left in the 1 TB drive that comes with it. It was easy to set up otherwise, and it took me mere minutes to get rid of Copilot and install Google Chrome.

Value

Though the iBuyPower RDY Scale B06 retails at $1,479, it is currently available for $1,379 (and it was $100 less than that last week with a code). That puts it much closer to the rigs we tend to see in our roundup of the best gaming PC deals every week. But even then, it is outpaced by the iBuyPower Element SE, which is $80 less and comes with a better graphics card (RTX 5060 Ti) and a better gaming CPU (AMD Ryzen 7 9700F).

Then, if you simply want an RTX 5060 rig, the ABS Cyclone Aqua offers that, plus 32 GB of DDR4 memory (more of it, but worse speed) and an Intel Core i5 14400F (another 10-core chip but with less cache) for $230 less. There are some compromises with this ABS build (as you will be making with pretty much any build right now), but its compromises are in good places.

As you might expect from iBuyPower, it has thrown a gaming keyboard and mouse into the box, and you get a month of Xbox Game Pass with it too. I'd say the peripherals are decidedly meh, with the Chimera KM7 membrane keyboard feeling a bit too squishy and weak under the fingers, and the wired mouse being rather slow. But for something free in the box, that's slightly above what I was expecting. And the keyboard even comes with a media dial, which is a slight plus. If I spent my entire budget on the PC, I'd be happy enough using the mouse and keyboard until I got together the budget for something better.

This is one of the hardest reviews I've ever written, because the market is ever-changing (for the worse), and this flux state means we have to accept less for more until at least 2027, and likely longer.

So, to simplify things, I thought I'd ask myself, 'Would you buy this rig today if you absolutely had to get one?' and my answer would be 'definitely not for its full price'. However, I have seen this rig go as low as $1,280, and for that price, it does put up a fight among the most budget rigs. In fact, it could even eke out the Element SE and its RTX 5060 Ti at the right time.

The only problem is that iBuyPower only seems to sell the RDY models via its own website, so you have less of a chance of picking one up at a big discount at Walmart or Best Buy. But giving more options for relatively reasonable prices is only a good thing in my book, and if it gives a potential buyer slightly more choice at the $1,000-$1,500 range, it's a solid enough addition to the market. iBuyPower has made some decent concessions for that price point; it's just a shame it has to do that in the first place

And it's very nicely made. The one we've got here comes in black, though there have been white variants of the 35L in the past, and the glass front and side panels allow you to see the glory of the RGB fans and CPU pump as they swish away. The mATX motherboard contains two RAM slots, both of which are filled, and there's a spare M.2 for an extra SSD if you don't mind removing the GPU to get at it, but the empty PCIe slot at the bottom is so close to the graphics card shroud you're not going to get much in there.

AMD says AM5 should be good for up to Zen 7, however, so there's scope for some upgrades here as long as you don't mind replacing all the RAM—it's 32 GB of 6000MT/s across the two sticks right now, so there should be some headroom.

On the outside of the case you'll be disappointed if you're expecting USB4—it's not here, and the Type-C ports top out at 10Gbps. Retro peripheral enthusiasts will be pleased by the four USB2 ports, though three of them are likely to remain unused by most users. There's just enough I/O here to make the 35L useful, with top-mounted ports that are easy to reach right next to the power button. On the back, there's no HDMI for the CPU's integrated graphics—it uses a USB-C's DP mode instead—but you'll be more interested in the trio of DisplayPort and single HDMI of the GeForce card anyway. This arrangement should at least prevent that embarrassing scenario where you plug your monitor into the integrated GPU by mistake and wonder why your games won't run nicely.

Previous

1 / 2

Next

1440p gaming performance

Testing our benchmarking suite at the PC gaming sweet spot resolution

Avatar: Frontiers of Pandora (1440p Ultra)Cyberpunk 2077 (1440p RT Medium)Baldur's Gate 3 (1440p Ultra)Black Myth: Wukong (1440p Very High)Black Myth: Wukong (upscaling) (1440p Very High)Metro Exodus: Enhanced Edition (1440p Ultra)Cyberpunk 2077 (upscaling) (1440p RT Medium)

Avg FPS

1% Low FPS

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

99

66

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

92.5

70

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

68

27

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

82

46

Avatar: Frontiers of Pandora (1440p Ultra) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	99 Avg FPS, 66 1% Low FPS
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	92.5 Avg FPS, 70 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	68 Avg FPS, 27 1% Low FPS
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	82 Avg FPS, 46 1% Low FPS

Avg FPS

1% Low FPS

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

74

54

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

72

56

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

66

55

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

66

52

Cyberpunk 2077 (1440p RT Medium) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	74 Avg FPS, 54 1% Low FPS
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	72 Avg FPS, 56 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	66 Avg FPS, 55 1% Low FPS
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	66 Avg FPS, 52 1% Low FPS

Avg FPS

1% Low FPS

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

132

65

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

79

56

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

134

53

Baldur's Gate 3 (1440p Ultra) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	132 Avg FPS, 65 1% Low FPS
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	79 Avg FPS, 56 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	N/A
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	134 Avg FPS, 53 1% Low FPS

Avg FPS

1% Low FPS

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

58

42

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

61

51

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

56

49

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

57

42

Black Myth: Wukong (1440p Very High) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	58 Avg FPS, 42 1% Low FPS
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	61 Avg FPS, 51 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	56 Avg FPS, 49 1% Low FPS
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	57 Avg FPS, 42 1% Low FPS

Avg FPS

1% Low FPS

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

103

80

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

92

71

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

83

51

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

83

66

Black Myth: Wukong (upscaling) (1440p Very High) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	103 Avg FPS, 80 1% Low FPS
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	92 Avg FPS, 71 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	83 Avg FPS, 51 1% Low FPS
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	83 Avg FPS, 66 1% Low FPS

Avg FPS

1% Low FPS

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

172

49

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

127

76

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

115

42

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

169

51

Metro Exodus: Enhanced Edition (1440p Ultra) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	172 Avg FPS, 49 1% Low FPS
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	127 Avg FPS, 76 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	115 Avg FPS, 42 1% Low FPS
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	169 Avg FPS, 51 1% Low FPS

Avg FPS

1% Low FPS

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

168

61

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

163

121

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

184

102

Cyberpunk 2077 (upscaling) (1440p RT Medium) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	168 Avg FPS, 61 1% Low FPS
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	163 Avg FPS, 121 1% Low FPS
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	184 Avg FPS, 102 1% Low FPS
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	N/A

System performance

Testing the core components of the PC with creator and synthetic benchmarks

Cinebench 2024Blender 4.2.0 (junkshop)7zip 24.07Handbrake UHD to 1080p conversion3DMark Storage

Single core index

Multi thread index

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

129

1283

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

131

1806

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

110

1046

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

128

1227

Cinebench 2024 Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	129 Single core index, 1283 Multi thread index
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	131 Single core index, 1806 Multi thread index
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	110 Single core index, 1046 Multi thread index
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	128 Single core index, 1227 Multi thread index

CPU (samples)

GPU (samples)

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

102

1958

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

103

2092

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

83

918

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

93

1909

Blender 4.2.0 (junkshop) Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	102 CPU (samples), 1958 GPU (samples)
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	103 CPU (samples), 2092 GPU (samples)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	83 CPU (samples), 918 GPU (samples)
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	93 CPU (samples), 1909 GPU (samples)

Compressing (GIPS)

Decompressing (GIPS)

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

124

143

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

132

154

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

115

131

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

115

131

7zip 24.07 Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	124 Compressing (GIPS), 143 Decompressing (GIPS)
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	132 Compressing (GIPS), 154 Decompressing (GIPS)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	115 Compressing (GIPS), 131 Decompressing (GIPS)
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	115 Compressing (GIPS), 131 Decompressing (GIPS)

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

89

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

80

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

75

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

85

Handbrake UHD to 1080p conversion Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	89
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	80
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	75
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	85

Overall index

Avg bandwidth (MB/s)

Access time (us)

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

2314

393.22

77

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

1764

299.97

101

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

2343

403.37

77

Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D

2572

441.63

70

3DMark Storage Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	2314 Overall index, 393.22 Avg bandwidth (MB/s), 77 Access time (us)
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	1764 Overall index, 299.97 Avg bandwidth (MB/s), 101 Access time (us)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	2343 Overall index, 403.37 Avg bandwidth (MB/s), 77 Access time (us)
Custom PC build | RTX 5070 Ti | Ryzen 7 9800X3D	2572 Overall index, 441.63 Avg bandwidth (MB/s), 70 Access time (us)

The games do run, however, and they run a little faster than the last Omen 35L PCG reviewed back at the beginning of 2025, which paired a Core Ultra 7 265K with an RTX 4080 Super and a larger 2 TB SSD. That PC was a little bit cheaper at the time, a difference of about 10% on the US dollar prices, with two years' worth of inflation and all this *gestures around wildly* to blame for the increase. You're getting an extra two frames per second (2.73%) in Cyberpunk 2077 at 1440p with upscaling off for your money.

That's assuming you upgrade your 35L every time a new one comes out, and most people probably don't do that. Though if you do, write in and the hardware team will refer you to one of their legion of therapists. For anyone coming from a RTX 20- or 30-series Nvidia card, however, it's going to be a major step up.

It's certainly a very quiet machine. The case fans spin all the time, though they do switch the RGB off on occasion, and all they're doing is feeding air over the motherboard heatsinks and to the GPU cooler, as the CPU is under a water block that sends its heat out the top of the case through a 240 mm radiator. Work it hard and all you'll hear is a rushing sound like wind through trees that doesn't stop, the working of the pump and the motors in the fans largely unnoticeable—of the PCG benchmark suite, it's only Time Spy Extreme that really gets it fired up, and only then near the end. It's a system that keeps everything very nice and cool in our tests, with even the idea of 70°C something neither the CPU nor GPU is prepared to entertain during a Metro Exodus run.

Thermals

Temperature of the core components during a stress testing game run

Gaming Temps

Avg CPU Temp (°C)

Max CPU Temp (°C)

Avg GPU Temp (°C)

Max GPU Temp (°C)

HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D

60

66

60

66

HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K

52

62

54

33

iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D

63

71

60

43

Custom PC build | RTX 5070 Ti | Ryzne 7 9800X3D

67

74

69

53

Gaming Temps Data

Product	Value
HP Omen 35L | RTX 5070 Ti | Ryzen 7 9800X3D	60 Avg CPU Temp (°C), 66 Max CPU Temp (°C), 60 Avg GPU Temp (°C), 66 Max GPU Temp (°C)
HP Omen 35L | RTX 4080 Super | Core Ultra 7 265K	52 Avg CPU Temp (°C), 62 Max CPU Temp (°C), 54 Avg GPU Temp (°C), 33 Max GPU Temp (°C)
iBuyPower RDY Element 9 Pro R07 | RX 9070 XT | Ryzen 7 7800X3D	63 Avg CPU Temp (°C), 71 Max CPU Temp (°C), 60 Avg GPU Temp (°C), 43 Max GPU Temp (°C)
Custom PC build | RTX 5070 Ti | Ryzne 7 9800X3D	67 Avg CPU Temp (°C), 74 Max CPU Temp (°C), 69 Avg GPU Temp (°C), 53 Max GPU Temp (°C)

Metro is always an interesting case study as it puts a lot of strain on a PC's CPU. The best results in testing for this game come from AMD CPUs, with the X3D models rising to the top, and so it proves here with the 35L's results coming within a whisker of an almost identically-specced custom system created for PCG's print magazine. The only desktop PC that beats the 35L (in this test, out of the systems PCG has tested) is an NZXT Player Three Prime that pairs an RTX 5090 with a 9800X3D, and which costs a lot more than the Omen 35L. In a game of diminishing returns, an extra 19fps for $1800 is well up there when you're already producing over 170.

This all goes to show why the AMD Ryzen 7 9800X3D is still PCG's favourite desktop processor, even though the AMD Ryzen 7 9850X3D now exists. Despite being built on AMD's latest architecture it's practically old-fashioned, having only one type of core, no NPU, and hyper-threading (SMT) enabled, yet trashes the competition in pre-built or DIY machines when paired with a suitable GPU.

Buying a pre-built desktop system like this is, in many ways, PC gaming on easy mode. You plug it in, it works, you get great results—there's no obsessive selection of parts, no creation of multiple lists, and certainly no screwdrivers. It may take a bit of the fun out of getting a new PC, but at least it's not a laptop.

The question of value will hang over the Omen 35L, especially with component prices the way they are right now. It's not a bargain system, but it sure is cheaper than something like the Maingear MG-1 Mk.II, and can at least cool its CPU effectively. What you get for that money hits pretty close to a price/performance sweetspot, though, and that's got to be worth a look.

1,000 Hz. That's 1,000 frames. Every. Single. Second. Welcome to the wild world of the new LG UltraGear 25G590B, the first ever 1,000 Hz gaming monitor.

Well, it's the first gaming monitor capable of 1,000 Hz at full 1080p resolution, as opposed to displays hitting that kind of refresh rate at, say, 720p. The tempting immediate reaction, of course, is to doubt the relevance and benefit of something so fast.

I mean, at what point do diminishing returns kick in with this refresh rate stuff? You could argue that anything above 144 Hz is overkill for most gamers. That was a common view just a couple of years ago. And it's certainly true that 144 Hz is enough for very smooth visuals and a good sense of immediacy and response.

But as the frame-rate ante has been upped in recent years, to 240 Hz, 260 Hz, 480 Hz and well beyond, that narrative has shifted. In fact, our resident competitive shooter, Jacob, is adamant that 360 Hz is the current sweetspot. Keen esports gamers, it turns out, really can feel the difference, especially in terms of latency, as the refresh rates ramp up. And so here we are in 2026, somewhat inevitably, with LG bursting through the 1,000 Hz barrier.

Aside from the headline refresh rate—the UltraGear 25G590B is actually capable of 1,100 Hz using an "overclock" mode—many of the rest of the spec list is, predictably, pretty pedestrian. That goes with the territory with this type of monitor.

So, we're talking 25-inch panel size and IPS technology. Brightness comes in at 400 nits, contrast is a mere 1,000 to 1 and there's only the most basic DisplayHDR 400 certification. So, this panel will decode an HDR signal, but it can't really render HDR visuals.

Again, all of that is expected in this class of gaming panel. It's not about high-DPI visuals or OLED-style per-pixel lighting. It's all about speed. Actually, it's primarily about one aspect of speed: latency. The other major metric of speed, pixel response, isn't such an obvious win for theUltraGear 25G590B.

LG has rated it at 1 ms for the most often quoted grey-to-grey transition. That's pretty quick for an IPS monitor, but not actually exceptional. Not when there are 0.5 ms LCD-based monitor models knocking around, never mind the OLED crew which all clock in at a mere 0.03 ms.

That's something of a surprise given the absolutely cutting-edge refresh rate. On paper, the 1 ms response perfectly matches the 1,000 Hz refresh. But then measures of pixel speed are infamously mercurial. There isn't actually an enforced, or even widely agreed upon metric for pixel response in the monitor industry. So, response ratings are really only ever a rough guide.

Anyway, connectivity includes two HDMI ports and a single DisplayPort input. Only the DisplayPort interface is capable of the full 1,000 Hz refresh and it's critical to note that you need a GPU that supports the latest Displayport 2.1 standard. DisplayPort 1.4 is only good for 850 Hz at 1080p.

In practice, that means for the full 1,000 Hz you need AMD GPUs from the Radeon RX 7000 series of 2022 onwards, while in Nvidia terms only the latest RTX 50-series chips will do. There are other system spec and configuration issues that you'll also want to consider if you're thinking about dropping a grand on a monitor this fast. But hold that thought, we'll come back to it.

Just to round out the speeds and feeds, further features include FreeSync and G-Sync support, Motion Blur Reduction Tech and small dash of configurable RGB lighting via the logo on the rear of the chassis. It's all housed in a pretty anonymous looking dark grey plastic chassis with fairly chunky bezels. But you do get a full complement of adjustment, including tilt, height, swivel and a touch of pivot, so it's easy to get this display set up just so.

All of which brings us to the most important question of the day. What is a 1,000 Hz monitor actually like to game with? It turns out, that poses quite the conundrum. Initially, it's easy enough to get a basic impression of the impact of 1,000 Hz on motion clarity. All you need do is load up something like UFO Test and watch those little flying saucers zip by at multiple different refresh rates.

And you know what? The benefits in terms of motion clarity do keep on coming, even as you climb beyond 500 Hz. Yup, 1,000 Hz is noticeably better, at least in that narrow, rather synthetic metric, when it comes to clarity.

But what about in-game motion clarity and, perhaps more importantly, latency? That's where the trouble comes. Because it's quite hard to achieve the 1,000 fps required to saturate the 1,000 Hz refresh actually in a game.

By way of example you cannot, to the best of my knowledge, achieve it in Counter-Strike 2, the CPU ends up bottlenecking performance long before that. Well, strictly speaking you can get that high, somebody managed 1,300 fps, but that was with an Intel Core i9 chip overclocked to 8 GHz courtesy of liquid nitrogen. Not exactly practical.

Something like an AMD Ryzen 7 9800X3D is probably the fastest stock-clocked CPU for running Counter-Strike. And that's good for about 600-odd fps at medium 1080p settings. I was getting in the high 400s with an Intel 14th Gen Core i9-based rig. So, yeah Counter-Strike 2 wasn't cutting it.

But there are other games which run fast. And that's how I found myself installing Valorant. The things I do for you people. But even Valorant wasn't giving me 1,000fps, more like 500 to 800 fps depending on how much is going on, on screen.

Actually, using this monitor raises a whole lot of questions about your rig and game engine setup. For instance, you want a mouse with a fast polling rate, but not too fast. An 8,000 Hz polling rate just saturates the CPU with hardware interrupts to no benefit.

Suddenly, everything from CPU cooling and memory timings, to enabling features like Nvidia Reflex become critical to eeking out every frame and minimising latency. Actually seeing the full 1,000 Hz in-game, though? That is far from straight forward even with titles where it may be theoretically possible.

With those provisos noted, what is 1,000 fps gaming, or as close as I could get to it, like? The immediate impression isn't exactly mind blowing. It doesn't look dramatically clearer or feel dramatically more responsive than, say, a 480 Hz panel.

But there is a subtle difference and the more you game, the more you key into it. I'm not sure the motion clarity advantage, such that it is, is really a major factor in that, however. Maybe a 1,000 Hz OLED panel would be different. But I think in subjective terms, motion clarity above 500 Hz on an LCD is likely a case of diminishing returns.

I think that applies doubly to a monitor like this that uses an IPS panel. It's a legitimate question to ask if the liquid crystals in this monitor are actually fast enough to fully deliver on the 1,000 Hz promise in terms of pixel response. Like I said, LCD response specs should be viewed with some healthy scepticism. And this monitor doesn't even claim to have the fastest IPS panel out there.

For the record, I'm not much of a fan of the Motion Blur Reduction feature, either, which is basically a backlight strobing solution. It dims the brightness and doesn't give much visible benefit at higher refresh rates. It is a bit more impressive at lower refresh rates, but then running at 120 Hz is not why you just spent so much money on this monitor.

The latency, however, now that's a little different. The sheer speed and immediacy as you whip your mouse around to the next target definitely gets under your skin. The more you play, the more you feel hard-wired right into the game engine, any and all sense of disconnection falls away. It's pretty cool.

Reverting to a lower refresh rate makes that a touch more obvious. Put it this way, once you've played for a while on a monitor that can respond fast enough for a game engine knocking out 700, 800, 900 fps and beyond, a 360 Hz panel feels like a tiny bit of a retrograde step. Maybe, just maybe even a 500 Hz display, too.

All that said, I also don't want to overstate the upside. You'd readjust to that "sluggish" 500 Hz monitor pretty quickly. And we are talking about very marginal benefits at best. Ultimately, then, this is going to come down to how serious you are about competition in games and esports.

In part that's because this monitor is pretty rough for everything else. Once upon a time, widescreen 1080p-ish monitors like this felt huge and futuristic. Anyone who was around when the Dell 2405FPW launched in 2005 will remember how awesome that panel was. Well, 20-odd years later, a 1080p 25-inch monitor now looks small and very pixellated.

The colours are fine, ditto the viewing angles. But this monitor is $999 / £899 and at that price point, you have so very many options. Heck, the monumental new 39-inch Alienware 5K2K RGB-stripe OLED I reviewed recently isn't really much more expensive.

Of course, the LG UltraGear 25G590B is unambiguously targeted at a very narrow niche of the most serious and committed esports fans and pro players. A 39-inch ultrawide OLED really isn't relevant. On the other hand, a 27-inch 1440p OLED probably is. That would mean a bit of a compromise on frame rate, but also a monitor that's a much better all-rounder for both gaming and general computing.

Indeed, it's a little hard to understand just where the money is going with this 1,000 Hz LG. Peruse our best monitor deals page and you'll find high-refresh 25-inch 1080p IPS monitors from about $100. There's an Alienware model with 320 Hz refresh for just $199. Yes, this LG will need extra driver ICs, a custom TCON, that DisplayPort 2.1 interface, a quick backlight driver for the Motion Blur Reduction feature. But none of those are hugely costly.

Ultimately, then, there's a pretty big tax being applied here in return for being at the very cutting edge in terms of refresh rate. If you want the absolute last word in frame rates and low latency, this is unambiguously it for now. If you're a pro esports gamer, for instance, the broader context of the monitor market really doesn't matter. For everyone else, you have to think there's better value elsewhere.

There has been no drought of good 4X games over the past decade. Part of that has been the steady work of Amplitude, whose Endless series has become a stalwart part of the strategy scene. Endless Legend 2 further cements Amplitude's legacy as a great 4X developer, delivering the studio's trademark creativity and standout visual design in a modernized package.

The original Endless Legend was a shot in the arm for a 4X genre that had gotten pretty stale. If you'd asked me in the years afterward if I wanted more Endless Legend, I'd have said yes. In the years since, many of its innovations—and those from sibling series Endless Space—have become standards in the genre. You might expect that to have dulled some of the charm. 12 years later, though, I still want more Endless Legend.

Sciences fantastical

I wanted more Endless Legend because there's never enough good science fantasy. Not the Star Wars kind, though that's fun too, but the kind where it's sword and sorcery with a technological flavor. Where it's mostly monsters and wizards. Bows, not blasters.

So I'm glad that we're back to Endless Legend's worlds of crashed starship survivors building robotic golems and battling with titanium swords and wielding magic that might just be nanotechnology. The setting and its weird people are just as good as the first time, though a bit more scientific-feeling. No dragons, for example, or elf stand-ins—instead we've got evolved desert lizards, coral robots, and sand spirit genies. It's another rich world that lets its developers break the rules of the genre in interesting and creative ways.

The new world of Saiadha is familiar in one important way: Like Auriga of Endless Legend, Saiadha is dying. This time though it's because the Endless, an ancient race, sealed something away beneath Saiadha when they terraformed it from a planet of desert dunes into one of storm-wracked seas and isolated islands. Now that something is coming back, which means the oceans are draining away into the planet's core.

Every few dozen turns a monsoon comes, shrouding the world in storms, and at the end of that monsoon season comes Tidefall: The shoreline recedes, unveiling sunken landscapes and ancient technologies that are now ripe for your faction to colonize and exploit—a perfect setup for a genre about expanding and building.

Aesthetics fantastical

Endless Legend 2 continues the tradition of the first game in another way with one of the most pleasing, coherent aesthetics in the entire strategy game space. The painterly color choices across terrain and units, both matte and gloss, make different areas of the world and the new biomes that emerge from the ocean very distinct.

Endless Legend 2 also makes very good use of texture on terrain to convey types of earth, sand, grass, stone, or even a field of vast living circuitry. My major critique is that the lighting is quite flat—it really only changes during monsoons—and something more dynamic or directional might have helped highlight the models and environments. Better to have simple lighting than effects that muddy the visual environment, though.

While it's still a best-in-genre user interface, interacting with Endless Legend 2 doesn't feel as aesthetically pleasing and crisp as the first game. The redesigned resource displays, menus bar, and action wheels are good, but the use of symbols on the map for cities and armies rather than clear, obvious text labels conveys less information while making the player do more work to understand what they're looking at. I had issues distinguishing the symbols for cities, armies, and resource locations until I learned where each was located. A temporary annoyance, but bad friction nonetheless.

The highlight for many will be yet another standout soundtrack from Arnaud Roy. Ethereal at times—subdued and ambient—it makes extensive use of a crystal organ. What most impressed me was how it changes as you go, avoiding repetitive loops and adapting dynamically to the screen you're on or what's happening in a battle round. Opening a diplomacy screen, for example, seems to dull out or soften the bass parts so you can hear the faction leader characters speak clearly.

Play fantastical

What may disappoint some about Endless Legend 2 is that many of the first game's signature innovations have become routine. Ideas like changing or terraforming the map, or your cities expanding over large regions, have been adopted by many others in the genre. The signature Amplitude resource system is present in full and simple to understand: Incomes for cities—FIDSI for Food, Industry, Dust, Science, and Influence—with tradeable luxury resource stockpiles for upgrades and everything else. Some might be bored of it, but it works really well, and I don't think you should mess with a good system.

What is still superb is how Amplitude wants every faction to break the rules in some way, repeating only a few tricks from Endless Legend or Endless Space. Particularly fresh is how Saiadha's receding oceans play into the genre tropes of a 4X.

The genre is at its best during two phases—the initial exploration and expansion, and the secondary squabble and race over who will control resources and which factions will actually fight with each other for them. Endless Legend's system of neutral territories is perfect for this, as you can stake claims by building camps, but must be wary of attack from enemies or marauding minor factions who can burn those down without declaring an official war.

The worst part of most 4X is when the map stagnates, empires get huge, and you engage in a grind toward whatever finish line you're aiming for. In Endless Legend 2, every time the seas recede and new land appears, the shorelines change—previously isolated territories may connect to each other, precious deposits or strategic resources may appear, and new dungeons for your heroes to delve or undiscovered minor factions may show up. Tidefall means your previously safe border may become vulnerable, or that your worst city is now part of a rich new frontier. In short, it throws you back into the most exciting and interesting parts of the 4X genre multiple times throughout a typical length game.

Tactics fantastical

The changing map also means your armies are in constant motion. You always want them moving from place to place, scouting out new territory, gathering curiosities, and either fighting or doing quests for neutral minor factions. At the same time you need them to guard your borders against potential aggressors, or against the terrible Doomwraiths that come each monsoon season, and you want them in position for the next tidefall. Where your military units and scouts are just map decoration in some 4X games, in Endless Legend 2 they're hooked into every part of the game loop in a way that feels like part of the exploration and wonder of a good fantasy story.

For all that, the tactical combat is perhaps less rich than I'd like it to be. The positioning elements of the turn-based fights are as good as ever, especially because of how it happens across the natural game map. You carefully position your armies to attack enemies from favorable terrain. As a faction like the Tahuks, who rely on ranged combat and want to exploit chokepoints, or the Sandshapers, who want to fight on desert terrain whenever possible, controlling space by positioning is still very good.

What leads to it feeling less rich is that it's very tightly wound—damage numbers are such that fights are nearly equal exchanges unless you outnumber the enemy 2:1, and scraping out an advantage often feels like exploiting the system by stacking buffs or abusing the terrain.

That said, 4X games are rarely well-tuned, but I found that the difficulty settings in Endless Legend 2 actually worked very well on their defaults. Higher settings actually delivered greater difficulty in consistent, measured steps, and the highest difficulties were so rough that you could end up in wholly unwinnable starts at times, which paradoxically I'd count as a real success.

That success, however, is still achieved by inflating the AI factions' abilities with bonus resources rather than making them better at playing the game. So sometimes the only way to win those tight tactical combats against a stream of armies you could never out-produce was by abusing the heck out of a system.

Weirdos fantastical

The factions are the highlight of this game, as with any Endless game, because of how asymmetrical they are. Each of them breaks or ignores some fundamental game mechanic.

Whether it's the Children of Sheredyn who'd rather build a fortress than a public plaza, or the Sandshapers, living mineral constructs who can't farm and must scrape together rare resources and terraform the world back into desert in order to gather the minerals they need to produce new population. There are even two "rogue" variants of factions that invert their primary playstyle—the defensive Sheredyn become an aggressive expansionist legion, the devouring Necrophage become an enslaving swarm of cyborgs.

There's less extremity than in Endless Legend in places—no faction can move its cities, for example—but there's still wild stuff in here that would be seen as unacceptable in other games. The friendly Aspects, a symbiosis of living corals and robotics, can spend their copious influence to force treaties like peace or trade on other factions. The Necrophage can only have one city, their hive, and are always justified in declaring war on others. (They're hungry, OK?)

There's just not another 4X where you can play as religious fanatic lizards who use a solar mirror laser to turn the world to glass. I appreciate that.

But I haven't always been sold on the Steam Frame. I spent the week prior to its arrival playing on the Pimax Dream Air, which includes the most spectacular Micro-OLED displays. That is a tough act to follow for any VR headset, let alone one with optics not dissimilar from the three-year-old Quest 3. As a long-term VR user and someone that's been privileged to play most of the best VR headsets in recent years, the Steam Frame felt too familiar an experience out of the box.

Over many weeks of testing, I've noticed more how the Steam Frame has eased a few of my fundamental gripes with virtual reality; things that cause me to leave my other headsets to gather dust on the shelf. Like how often I need to set up my playspace, set the floor height, peer out from my headset at my computer, adjust the strap, or untangle myself from a cable.

A wireless dongle included with the Frame takes care of the cable, delivering a seamless and reliable connection between it and a gaming PC. The fully-featured SteamOS means the first step upon donning the headset is choosing a game, and the second step playing it. The included controllers work with every game, VR or non-VR, and are tracked with pinpoint accuracy. And with such a lightweight design, this is one headset I'm not in a rush to take off.

I want to use the Steam Frame more than any other headset I own right now. That includes some that are double the price. Ack, I've mentioned money. Now I have to do the hard thing and bring this spiralling back down to Earth: The Steam Frame starts out a $1,059 for the 256 GB model. I haven't spoken to anyone, even Valve employees, who has much positive to say about the price. The memory crisis has done a number on the Steam Frame, as it did the Steam Machine.

The Steam Frame isn't a cheap way into virtual reality for the masses. It has to be viewed as an enthusiast offering; or, at the very least, you have to care a bit about VR to purchase one. Those of us wanting to try VR for the first time are still better off with a Quest 3 or Quest 3S for value for money; and those of us that already own a Quest 3—that's a lot of us—it's a tough upgrade, lacking any major visual overhaul. There are undoubtedly better options for enthusiasts looking for that, like the Bigscreen Beyond 2. Yet it is far more convenient than high-end headsets that still rely on Base Stations or heavy tethers.

Ultimately, I believe VR to be a holistic experience. Focus all your efforts on one aspect of a headset and the user experience is likely to suffer moreover for something else; the weakest link. Take it from me, even high-end optics fall flat after the umpteenth time resetting your playspace due to poor tracking. The Steam Frame does, then, have its virtues: It's by far the best all-round VR headset for user experience, with no major faults besides the price; it's just that price matters more than most.

Steam Frame - Gallery

Steam Frame - Features

Steam Frame - Optics

The Steam Frame's optics are, on paper, the least exciting thing about it. We've seen a handful of headsets with similar configurations, from the Quest 3 to the Pico 4 Ultra; neither of which are brand new.

Graphical fidelity feels much the same on the Steam Frame as the Quest 3. There's an amount of aliasing (jaggies) on straight lines and sharp details, which adds a touch of flicker as your head moves around. There is also a slight hint of chromatic aberration to the edge of some brighter details, such as the Steam Frame logo upon startup, though this is quite minor.

Text is clear to read when scaled appropriately, though can get tougher to make out when it's outside of the fairly large focal area of the pancake lenses or at a greater distance. Similarly, some details in game worlds can be lost when viewed from far away, and objects with meshes can get a little distorted or flicker—something that higher-end headsets, like the Pimax Crystal Super or Dream Air, don't suffer from. Those two headsets are available with Micro-OLED displays, which offer much more vibrancy and range of colour than the Steam Frame, but for huge price tags.

The Steam Frame offers, more or less, a level of quality that I've become accustomed to in relatively affordable VR headsets.

Where the Steam Frame outperformed my expectations is in binocular overlap. This is the amount of an image (given as a percentage) that is covered by both left- and right-side displays at the same time. It's not as often talked about as resolution and refresh rate, but it's hugely important for long-term comfort while wearing a VR headset. The Steam Frame has a pretty high binocular overlap, according to my own testing in the 'What is my Field of View?' application (more commonly called WimFov). This is with an IPD set to 64.

• Steam Frame - 77%
• Meta Quest 3 - 69%
• Pimax Crystal Super 50 PPD - 77%
• Pimax Crystal Super Micro-OLED - 86%
• Pimax Dream Air (standard FOV) - 86%
• Pimax Dream Air (wide FOV) - 77%

The Steam Frame is right up there with the best of them for binocular overlap, beaten only by the Micro-OLED optics used in both the Crystal Super and Dream Air. The Quest 3 has a lower level of binocular overlap, which can lead to discomfort or worsened 3D effects, though it does depend on your IPD and various other biological factors. I tend to struggle with headsets with a low overlap over time; they strain my eyes.

Binocular overlap, as measured in WimFov

Valve is pretty honest when it comes to the FOV. The Steam Frame's horizontal field of view (HFOV) is noted as 'up to 110 degrees', and lo and behold, I measured it at that. That's not as common as you might think. I often find the effective HFOV to be lower than the stated FOV. For example, all three of the headsets in the table below have a listed HFOV of 110 degrees.

There are a range of factors that can have an effect on effective FOV, not the least bit my own biology. Yet one big factor is the design of the padding between one's face and the lenses. Valve seems to have nailed the padding design to maximise the effective FOV; thick enough as to not be a detriment to the long-term comfort yet ensuring a one's eyeballs are right up to the lenses. Using the glasses spacer included with the Steam Frame will knock this back a touch.

Beyond the numbers, the Steam Frame's optics plonk you right into the game world. There's no visible distortion until the very extremes of the lenses, and even then it's not that bad by comparison to some. The gap around the nose is a little noticeable during gameplay, and made worse by the glasses spacer, but you have to be kinda looking out for it to notice it too much. That can be reduced with the extended light blocker in the Ergonomic Accessories Kit for an additional $59 (89 CAD / 69 EUR / 59 GBP / 99 AUD).

There is a degree of glare, as I find with most headsets with pancake lenses, but to a lesser extent than the Quest 3. It's only really noticeable when there's a very high level of contrast being displayed, which isn't all that often in my experience.

For the most part, when I'm trying to throw barrels, bottles, and cans at headcrabs in Half-Life: Alyx because I've run out of ammo, I'm not thinking about the FOV, binocular overlap, or glare; I'm thinking, 'Ah, ah, ah! No, no, please, just die already! Oh, god!'

That's positive, right?

Steam Frame - Performance

The best way to play on the Steam Frame is over the included wireless adapter. It's a USB Type-A dongle providing a 6 GHz connection beamed directly at a dedicated antenna on the Frame. Another antenna on the Frame connects directly to the local Wi-Fi. The Frame actually taps both connections for game streaming from a local PC; selecting between the two in real-time for the lowest packet loss and most robust connection.

I'm yet to notice any connection issues over the dedicated dongle, whether in my own home or in our busy office. Nor any visible dips in quality for the game being beamed across. That's really impressive, as Valve's is using what it calls Foveated Streaming to reduce the bandwidth requirements over the wireless connection. It does so by only streaming the game in the highest detail in the area where one's eyes are looking. The rest is streamed at a lower quality level. The Frame's built-in eye tracking has to keep up with your pupils to ensure a smooth experience—and it does a superb job of it.

The Frame supports a refresh rate up to 144 Hz, though it defaults to 90/72 Hz, depending on the power mode selected. In the battery-saving Playtime mode, it will only run at 72 Hz, which is pretty passable in my opinion. At the expense of battery life and processing power, it can run anywhere from 90–144 Hz in the settings menu.

The Qualcomm Snapdragon 8 Gen 3 (SM8650) chip does a pretty good job of rendering games locally. It does, however, require gentler programming in the graphics menu.

Take Half-Life: Alyx, for example, as Valve has created a version of the game that runs natively on the Snapdragon chip. It is of much lower quality than you'll experience by streaming the game from a local PC over the provided dongle. Valve has added foveated rendering to the game (dynamically changing the rendering level based on where one's looking) in order to improve performance. Unlike foveated streaming, foveated rendering is far more noticeable in my peripheral vision.

Though Half-Life: Alyx running at all on the Snapdragon chip is impressive, as are other games that are able to run just fine after being chopped up by the two compatibility layers, Proton and FEX. Proton deals with Windows to Linux; FEX deals with x86 to Arm. I tried out a few VR and non-VR games standalone on the Steam Frame and it often dealt surprisingly well with them. These included a few Steam Frame Standalone Verified games and one that hadn't yet been checked.

• Eleven Table Tennis (verified): smooth, high fps
• Marvel's Spider-Man Remastered (not verified): failed to run past splash screen
• Escape Room VR (verified): smooth, decent fps
• Hades (verified): smooth, high fps

Marvel's Spider-Man Remastered is Steam Deck Verified; however, that doesn't mean it will run on Frame, as it using FEX as well as Proton. Essentially, there are a few more hoops to jump through.

Hades, on the other hand, runs great via both translation layers. It's likely a boon to the Frame that the game has a Vulkan version already available. If I understand it correctly, Vulkan should let Valve skip some of the translation work and improve performance. The game reached around 100 fps, so it definitely didn't seem to suffer for any of the funny business in the background. It's best to set the Steam Frame's cap to beneath this figure, anyways, as it keeps the visuals steady and stable. Once I set the system to 144 Hz, which it couldn't reliably hit in Hades, moving my head would cause the game screen to shudder.

Eleven Table Tennis is a great reminder of when low latency and high refresh really matter in VR. I've been trying to beat the silver-level AI in that game for weeks now. I had, at times, thought it impossible, especially after trying and failing on three different VR headsets (Frame, Dream Air, Quest 3).

Installing the game on the Frame and running it locally at 120 Hz, with the ergonomic straps removed from the controllers to aid my front spin, I finally cracked it. I nailed some fast return balls that the stupid cat (the AI has a cat's face, don't ask) couldn't return. It was a silky smooth experience, even on the Frame's internal 'Turnip' Adreno GPU. I feel the local processing and high refresh really helped me here; but it drained the heck outta the battery on this thing: 50% gone in under 30 minutes.

Nevertheless, the standalone functionality feels less crucial to me. I will almost always stream games from my gaming PC via the wireless connection, due to the performance, quality, and battery life improvements I'll get onto shortly.

Steam Frame - Ergonomics

The Steam Frame weighs in at 440 grams. On my scales, it's more like 403 grams, but that's more a criticism of my crappy scales than anything else. What's important is that it is exceedingly light; the second-lightest from my own measurements.

• Steam Frame: 403 g
• Quest 3: 437 g
• Quest 3S: 454 g
• Crystal Dream Air: 280 g
• Crystal OLED: 848 g
• Crystal normal: 837 g
• Sony PS VR2: 499 g

Comfort in a VR headset is not all about weight, but in this case, the Steam Frame's small stature helps a helluva lot. Valve has balanced the weight of the Steam Frame across the front and back of the device, with the battery as a counterweight to the headset's 'core module', containing the optics, SoC, and most other technical bits. With the weight of the battery elsewhere, the core module weighs only 157 grams. Valve has that at 185 grams but hey ho; crappy scales.

Ample cushioning around the face and on the rear of the battery help to alleviate most discomfort from the weight of the device during play sessions. Unlike the Quest or older, heavier headsets, the weight does not feel like it is localised to your cheekbones or around the eyes. Even after hours of play to test the battery life, I didn't notice any patches of red or indentation lines on my face.

If you wear glasses, the Steam Frame is pretty accommodating. There's a spacer included in the box that adds another 4-7 mm between your face and the lenses. It's the easiest to add into the headset that I've used to date, too. It's all magnetically attached—the padding and glasses spacer can be replaced in no time at all.

Ergonomics accessories kit

Valve is offering an Ergonomic Accessories Kit as a separate purchase alongside the Steam Frame. It costs 59 USD/89 CAD/69 EUR/59 GBP/99 AUD and includes three premium upgrades:

• Controller straps w/ compatible battery doors
• Extended light blocker
• Top headset strap

I reckon most Steam Frame customers would buy the Ergonomic Accessories Kit without question. I probably would, too. But I've done my duty and played extensively with it fitted to the headset and without, and I have to say, it's really not a must-have.

The headset doesn't move around much with or without the added top headset strap. It is, however, useful if you're planning to pass the headset around a large group of people, as it makes it easier to adjust from person to person in a dash. I suppose if you're playing Beat Saber or anything a little more active, you may want the added support, but again, I don't think it's all that necessary.

The extended light blocker does a pretty good job of keeping light out from your nose, and that's especially true if you wear glasses like me; however, I've played mostly without and it's not been a detriment to my time in VR.

Then there are the most contentious upgrades of the three: the controller straps. These replace the cord straps included with the controller with straps that tighten around the entire hand, akin to the Valve Index design. Once installed, I can make better use of the finger sensing features of the Steam Frame controller (capacitive buttons combined with some sensing of the lower digits) and they're easier to slip into than the standard cord option.

On the flipside, the Frame's controllers are not as elegantly designed for this sort of strap as the Index's. I had a few members of the PC Gamer team try out the Steam Frame one by one in the office, and on numerous occasions I received the feedback that they cannot comfortably reach all the buttons. Take away the straps and that issue goes away.

In my experience, the straps are an improvement for games where grip plays a large role, such as Half-Life: Alyx. I can launch an item towards me and catch it naturally without thinking twice. Whereas without the straps, I had to sorta relearn to lift and grab, as two separate movements. But I definitely don't mind using the controllers without the straps, and for playing 2D games, the more natural position without the added straps feels best.

Steam Frame - Controllers and tracking

The tracking on the Steam Frame is excellent; perhaps even its strongest feature.

It's inside-out tracking using four monochrome cameras, combined with IR illuminators for low-light conditions. Those monochrome cameras aren't the best for passthrough or XR play—they offer a rough blue fascimile of the real-world behind the headset—but they work great for mapping the movements of the headset and controllers. That means no more Base Stations, as were required with the Valve Index; just the Frame itself keeping track of its position in VR.

What impresses me most is how consistent the boundaries of the playspace seem to be when using the Frame. It's yet to lose track of a boundary, floor height, or its own location within that space in my many weeks of testing. Whether I place it on the floor, move it around, turn it off and on again, or even move furniture around it, the playspace is locked in.

The headset does a great job with the controllers, too. The pair provided with the Frame support 6DOF and an inertial measurement unit (IMU) to correctly orient themselves in the world with excellent accuracy.

I'm yet to experience a single moment where one or both of my virtual hands go wandering off into the distance. Neither have I experienced what I will shamelessly call a floppy hand problem, where my virtual hand feels stuck on some invisible object and flops around like a fish as I try to wiggle it free. I have had both of those happen to me in the past couple of weeks with another headset, the Dream Air, but none so far with the Frame.

The Frame's controllers are similar in shape and style to those included with the Quest 3/3S. That's no bad thing; both are comfortable and intuitive. Yet Valve has also added more buttons to its latest pads, including ABXY on the right controller, a d-pad on the left controller, and bumper buttons to both.

These extra inputs aren't required a whole lot in VR; most devices don't make use of them. They are instead useful for playing 2D games on a virtual screen within the Frame. I've been testing this out with the original Hades: dragging the screen to a massive cinema-like size and putting my gaming chair into its deepest recline. It's pretty good, and even though I think the novelty of playing on a virtual big screen is already beginning to wear off, I can see why Valve says you can slip the Frame on and then decide which game to play. It can pretty much play whatever from your library with minimal fuss and without switching controllers.

For the record, you can pair up an external controller, keyboard, or mouse to the Frame over Bluetooth, including a Steam Controller. I haven't found much need to do so as of yet.

Steam Frame - Software

SteamOS is going places. Once limited to the Steam Deck and whatever one dared to load the ISO onto, it has since made its way into the living room and other handhelds. With the launch of the Steam Frame, the Arch Linux-based operating system breaks into the third dimension.

The Steam Frame offers a familiar interface for any PC gamer. It's just SteamOS, or Big Picture Mode, except its hovering in front of you. It has extensive menus for adjusting various settings on the Steam Frame, a status bar with some handy functions for VR, and a slightly tweaked interface for moving and enlarging windows within your expanded field of view. Otherwise, it's just Steam.

All your games in your library move over to the Steam Frame upon signing in. There are new categories automatically added to the interface to separate VR titles from non-VR titles, and in some cases, games are divided by those installed on the headset and those that can be streamed from a nearby PC. Once you select a game, you can click the arrow to the right of the green play button to select between playing it locally or streaming it from your gaming PC; depending on if it is installed on one or both devices.

Valve also offers the Steam Frame Standalone Verified program to give an idea of whether a given game is going to play nicely while running on the Frame's own hardware. Few games are built for the Frame—as far as I know, it's just Half-Life: Alyx—most are going to be either Android games running through Lepton, Valve's Android container, or use translation layers, Proton and Fex, in order to be made to run on Linux and Arm.

The Steam Frame wouldn't be around were it not for SteamOS, Proton, and FEX. Meta has Horizon OS, Google has Android XR, and Apple has visionOS for their respective headsets. Only Horizon OS feels in any way a fit for gaming, but even so, Meta has shifted to a more XR-centric interface and put a hold on third-party headsets using it. So that's out. That really means Valve had no other option than to do it itself with the Steam Frame.

And thank God for that. The biggest selling point for of the Steam Frame and SteamOS in virtual reality for me is that Meta has nothing to do with it. While you no longer need a Meta account for the Quest 3/3S, you still have to use the services of a company with big AI aspirations and a problematic privacy and safety record.

As for games available for the Steam Frame, it has plenty to choose from already on Steam, and it's set to benefit from Valve recently adding support to Steam for Android games. The Steam Frame also comes with a free copy of Half-Life: Alyx, and I can think of no better game that this to show off what it can do; able to both be streamed from a gaming PC and run locally on the Steam Frame's Snapdragon processor. It's the best VR game around today, despite being released over six years ago.

That touches on one issue facing VR in general today: the availability of new games and experiences. Plenty of the best VR games are the same ones we have been playing for many years, such as Beat Saber, Phasmophobia, Skyrim, Blade & Sorcery, Job Simulator, Arizona Sunshine; which I've pulled from Steam's own popularity rankings, among a sprinkling of newer titles. One newer title I've really enjoyed in the past year is Ghost Town, made by Fireproof Games, but even it is a year old now.

While Steam itself sees ever-increasing numbers of games published to the platform every single year, virtual reality games published on the platform have plummeted since an initial surge around 2017-18. Meanwhile, the Meta Horizon Store appears to still be trending upwards each year, though I can't be sure how many of those are actual games. A lot seem to be training applications.

If Valve can attract new developers to share their existing and future Android games on Steam, perhaps we will see a renewed surge of VR games on the platform. As of right now, it feels like fresh experiences are few and far between.

Steam Frame - Battery

You don't have to constantly worry about tripping over a cable with the Steam Frame. Instead, you have to be aware of your battery level.

The Steam Frame comes with a 21.6 Wh Li-ion battery, located at the rear of the strap. It has a USB 2.0 Type-C port off at an angle from its left-hand side, which is used pretty exclusively for charging the device due to bandwidth limitations for data.

There isn't an adapter included with the headset. Instead, it can be charged with any suitable power adapter up to 45 watts.

Valve is selling its own adapter for $29 (49 CAD/39 EUR/29 GBP/49 AUD). While I'm personally not too bothered about the lack of charger in the box—I have plenty already and I'm all for cutting out excess chargers—some of my colleagues feel it's a bit of a miss at this price point.

It takes a while to fully charge the Steam Frame from flat. In one test, using a 140 W Anker power adapter, it took 151 minutes to go from completely empty to fully charged. In another, using Valve's own power adapter, it took 142 minutes. So a good two hours and change for a full charge.

Here are the battery levels and times recorded from my testing:

Using a power meter on my wall outlet, I recorded the power draw in both tests around ~ 11 W at all times.

I asked Valve about the battery charging behaviour ahead of launch. They told me:

"An empty battery takes around two hours to charge to full capacity. However, the charging is not linear. There is fast charging from 0 to 65%, and regular charging from 65% and up. Charging to 65% is expected to take less than 40 minutes. (Note: these values may be optimized over time, and are therefore subject to change.)"

I hadn't seen any sign of fast charging to 65% in my own tests at the time I reached out; however, prompted by Valve's response, I tried one last test. I ran the battery down to near-flat, and while the headset was still turned on but not playing any games, I plugged in the charger.

This time, the power meter read ~31 W—three times the draw I recorded when the headset was powered off. When it reached 65%, it dropped to ~18 W.

I have felt a little impatient with the Frame on a few occassions—a feeling of having to wait too long between sessions once the battery had died. Though I did improve things by not letting the Steam Frame die completely between charges, using a suitably powerful power bank in my pocket (around 25 W should do it), and using the Frame's various power saving modes.

Depending on the mode of use, the Steam Frame can last between one and two hours of play time:

• Half-Life: Alyx (Arm version, running locally): 63 minutes
• Half-Life: Alyx (streaming from PC w/ Frame set to Playtime mode): 114 minutes

So, for every hour played in a game running on the headset, you'll be more or less charging for two; however, the longevity of the headset while streaming from a PC is pretty impressive.

I took these figures with the Frame set to Playtime mode, which favours battery life over performance and limits the frame rate to 72. You can push higher performance at a cost to longevity.

I noted the power draw of the headset in each mode while playing Half-Life: Alyx. When the game is running natively on the Steam Frame in standalone mode, the whole system draws around 17-19 W. Instead beam the game from my gaming PC using the included dongle and the Frame draws as little as 10 W. Nearly half as much power draw; nearly double the battery.

There may be another option for extending the Frame's battery life in future, too. Valve has confirmed to us it is working on a new strap with hot-swappable batteries and BMR drivers to match the excellent speakers on the Valve Index. This will be a swap-in upgrade for the modular Frame. No launch information for that one just yet, though.