AMD Radeon VII Review: This Isn't the 7nm GPU You're Looking For

This site may earn chapter commissions from the links on this page. Terms of apply.

The AMD Radeon 7 isn't a GPU we always expected to see as a consumer production. AMD repeatedly indicated information technology had no detail plan to bring a 7nm to this infinite, preferring to keep its first run of 7nm hardware reserved for servers and the AI/ML space. And so, at CES 2019, AMD CEO Lisa Su appear AMD would bring a new GPU to market to compete head to head with the RTX 2080 at the $700 toll point. Fast forward to the present day and hither we are.

Vega-vs-RadeonVII

The Radeon VII is based on the same 7nm silicon as the Radeon MI50 and MI60. AMD shrank the die significantly between the ii GPUs SEEAMAZON_ET_135 See Amazon ET commerce and used the space savings to clasp in another gear up of HBM2 chips, doubling bachelor RAM bandwidth. If you intendance near sheer memory capacity in your graphics card — and to exist clear, in that location are professional and scientific applications that benefit from large GPU buffers — the Radeon VII is the only carte du jour on the market that gives you this much RAM for under a thousand dollars.

Radeon-VII-Table

Boosted FP64 Performance

The Radeon 7 has another reward over other cards on the market, though AMD kept this i tight to their vest. Until today, the highest FP64 functioning y'all could purchase in a consumer GCN GPU was the Radeon 7990, a short-lived dual-GPU product that's nearly v years old. Initially, AMD communicated that its Radeon 7 GPU was capable of just 0.88 TFLOPS (1/16th of FP32 operation). Instead, the Radeon Seven is capable of a whopping 3.46 TFLOPs, or just under one-half of the MI60's maximum functioning (one/4 FP32 performance total).

This significant repositioning puts a new spin on the Radeon Vii, positioning information technology more than clearly for double-precision compute workloads. Unfortunately, this news wasn't communicated until right before the review NDA, which means we won't have fourth dimension to take a particular exam of the GPU in that context. We intend to write a split article investigating compute and scientific workloads on the Radeon Seven, which will give u.s.a. an opportunity to examine this side of the GPU more thoroughly. Today's review volition focus on the consumer side of the equation.

As with the first 7nm GPU, all eyes are going to be on the Radeon Vii to see how well it performs on the procedure node. We already have some information on this point. Every bit the table to a higher place indicates, the Radeon 7 dice is ~67 percent the size of the Vega 64 die. Clock rates take improved only modestly, with the base of operations clock up by 1.09x and the heave clock increasing by 1.13x.

Most of the improvements between Radeon Vega and Radeon VII boil downward to the huge increase in RAM bandwidth and total VRAM buffer. The clock gains from the 7nm shift are very small. AMD did note to us that information technology had reduced the GPUs internal latency and made a few performance-enhancing tweaks to the architecture, simply information here was limited. Those hoping that AMD would have a new major AI or automobile learning initiative to denote — something akin to DirectX ray tracing support or new antialiasing methods — will be disappointed.

New Thermal Monitoring

The Radeon 7 uses a new type of thermal monitoring system rather than the old, edge-mounted GPU thermistors it had previously deployed. Radeon Vii has a total of 64 temperature sensors mounted across the die, 2x the number of Vega 64. Going forward, Radeon VII GPUs will employ the maximum temperature measured beyond the dice, known as the Junction temp, to control GPU behavior. According to AMD: "Controlling based on Junction Temperature from the extensive sensing network allows each GPU to reliably maximize its performance potential while reporting an boosted temperature that is more than representative of the hottest parts of the GPU."

AMD notes that Junction temps volition be higher than what gamers are used to seeing in the past. This shouldn't exist considered a problem, and the GPUs are designed to concord these temperatures safely. Junction temps of 110C are not unusual or considered problematic.

The Story So Far…

Before we dive into performance figures, let's revisit how the GPU market has evolved in the by 6 months. Nvidia'due south RTX refresh cycle this past fall didn't do much to amend functioning-per-dollar. Of its new high-end GPUs, simply the RTX 2080 Ti genuinely moved the ball forward on functioning. The RTX 2080 and RTX 2070 are slightly faster than the GTX 1080 Ti and GTX 1080 respectively (recall eight-12 per centum), but carry higher prices than their predecessor GPUs did. Nvidia's justification for these price increases has been to indicate at its new ray tracing feature as justification. ExtremeTech historically takes a very dim view of ownership hardware for features you tin can't use, for reasons we explored in-depth as part of our RTX 2080 and 2080 Ti review.

But therein lies the rub. Fifty-fifty if y'all concord that Nvidia's RTX engineering science is a risky bet, AMD hasn't baked anything equivalent into the Radeon 7. In the past, one GPU vendor sometimes zigs when the other zags — Nvidia went all-in for iii-D spectacles and monitors a few years back, while AMD threw its weight behind Eyefinity and multi-monitor gaming. In this case, RTX and DLSS were major launch features for Nvidia, while AMD'southward major launch feature is the straightforward promise of more performance relative to its previous generation of graphics cards.

Gamers hoping that AMD would bring a GPU to marketplace that shook upwards the condition quo or at least forced Nvidia to lower its prices are going to be disappointed (once again). AMD is targeting the RTX 2080's performance at the RTX 2080's $700 price betoken using Vega's 300W power envelope.

Test Configuration

All testing was done on an Asus Prime Z370-A using an Intel Core i7-8086K with 32GB of DDR4-3200 and the latest version of Windows x. A Thermaltake Toughpower lxxx Plus Titanium 1250W PSU was used for testing.

All of our results can exist parsed in the slideshow beneath. Click on each graph to open it in a new window.

Functioning Summary

AMD promised that the Radeon 7 would exist capable of tackling the RTX 2080 SEEAMAZON_ET_135 See Amazon ET commerce , and that'south more often than not what we meet here. In our complete suite of tests, the Radeon VII offered 96 percent the performance of the RTX 2080. That's within a 5 percent margin of error, and shut enough to declare that yes, the two solutions are by and large competitive.

The Radeon Seven tends to showroom superior scaling to the RTX 2080, which is to say, it tends to lose less performance than its competitor as resolution increases.

Ability Consumption, Estrus, and Racket

Note: My GPU exam rig uses a V3 Voltair CPU cooler, which includes a Thermo-Electric Cooler, or TEC. This consumes additional electricity. Every bit a consequence, our idle and ability consumption figures may exist higher than elsewhere on the web. Running 32GB of RAM in XMP at DDR4-3200 as well substantially increases ability consumption compared with stock voltage and DDR4-2400.

Our ability consumption figures are taken from the tertiary loop of a Metro Last Calorie-free Redux criterion run at 1920×1080. All detail settings adjust to those used for our GPU reviews.

PowerConsumption-Metro-LastLight

AMD's absolute power consumption has scarcely budged compared with Vega 64, merely its functioning per watt has improved significantly. Using the automated undervolting option on our Radeon VII decreased power consumption by ~7.5 percent without harming frame rates at all.

One time you lot factor in the Radeon VII'due south increased performance, the GPU is indeed significantly more than efficient. The Radeon VII consumes roughly 75 percent as much power every bit the Vega 64 per frame of blitheness drawn. Actuate its underclocking feature, and this drops to 70 percent. Merely the RTX 2080 consumes simply 63 per centum the power of the Radeon Vega 64. AMD's 7nm GPU draws roughly the same amount of power equally its Nvidia rival, but it isn't quite every bit efficient on the whole.

Finally, there'southward noise. I don't own a dB meter, but folks — Radeon VII ain't tranquillity. Overall, it's comparable to the Vega 64, but there are moments when the fans on the Radeon Vii kick harder. They as well tend to ramp upwards faster. Anybody has their own personal tolerance for this sort of thing, but I consider the noise profile of these cards to be a significant negative.

At this point, the racket situation has become ridiculous. Always since at least Hawaii, reviewers have hitting AMD for the noise contour of its reference designs. To its credit, the company has at least attempted to address this, but its most high-profile attempt to set the problem with a water libation created an even bigger mess. Vega 64 was a loud GPU, louder than I'k personally comfortable installing in my own system. Radeon VII doesn't ameliorate on this at all. At this point, it'southward articulate AMD doesn't actually take whatsoever interest in edifice or outfitting its reference cards with coolers that match the operation of what Nvidia ships (and what Nvidia ships isn't always great, either, mind yous). Hawaii launched over five years ago. Why are we still waiting for AMD to actually fix this in a high-cease GPU that isn't the Radeon Nano?

I expect a $700 GPU to accept a amend noise profile than the $12 box fan I bought at Aldi.

Conclusion

The Radeon VII is a rough match for the RTX 2080'south functioning. Information technology also lacks one of the RTX 2080'due south specific liabilities. When Nvidia launched the RTX family, it asked customers to consume a significant price hike relative to the company'south and then-current GPU lineup. AMD isn't making that fault. The Radeon Seven offers more performance than whatsoever AMD GPU before information technology. It's 1.33x faster than the Vega 64 with no other pesky last-gen cards to muck up the stack.

Of course, given that the Vega 64 currently sells for $400, you're buying ane.33x more performance for one.75x the money. That'southward exactly the opposite of the kind of ratio we prefer to see. It's hard not to wonder what 7nm Vega'southward performance might have looked like if the GPU had fielded more than ROPs and texture mapping units to accompany its enormous bandwidth increase — again, nosotros doubtable that consumer games but weren't the primary market for this carte du jour and that its overall design reflects that fact. We intend to write a divide commodity focusing on GPU compute and will report dorsum with how the Radeon VII'southward boosted memory bandwidth aids it in those contexts.

BoxFront-and-GPU

For me, the bottom line is this: I have consistently argued that Nvidia erred in raising prices when it launched the RTX family. I accept argued that there were likewise many reasons to believe RTX features won't be particularly useful during Turing'south lifespan to justify buying into the family, particularly when older GTX cards based on Pascal offered such competitive performance.

Now we have the Radeon VII. It lacks RTX/DXR features just doesn't offering an culling. Its larger HBM2 buffer and interposer costs may explicate its pricing — I've wondered earlier if Nvidia's determination to enhance prices would actually make it make sense for AMD to bring 7nm Vega to the consumer market — merely they don't arrive a particularly proficient value. How much value do I put on ray tracing right now? Non much. Only a feature that'due south merely useful in a tiny number of games is however at least arguably more useful than no feature at all. Nor is in that location much reason to believe that the 16GB HBM2 buffer will go a workout any time soon. Game developers will always target their titles to the GPUs that people practically ain; having one consumer GPU with 16GB of onboard memory isn't going to move the needle in terms of future game VRAM requirements.

If AMD had managed to bring Radeon VII in at $600, it would take a genuine value argument to brand relative to Nvidia'south production stack. If it had outpaced the RTX 2080 at the same price, it could argue for superior rasterization performance with higher dissonance levels every bit an acceptable trade-off. Instead, what we have hither — at least in the consumer marketplace — is a loud RTX 2080-equivalent without the admittedly dubious features Nvidia tried to use to justify its price increases.

You know what'southward worse than an RTX 2080 with dubious features and a bad price point? A loud RTX 2080-equivalent with no new features at all and the aforementioned bad price point.

The story isn't all bad here. AMD was able to take advantage of its shift to 7nm to improve its overall competitive standing against the RTX family, and the Radeon Vii competes against the GTX 1080 Ti / RTX 2080 more than effectively than Vega 64 fared against the older GeForce 1080. Just I can't hammer Nvidia for months over the toll increases and positioning it introduced with Turing but to turn around and laud AMD for delivering a GPU that roughly matches on perf just offers fewer features and college racket, uncertain as the value of those features may exist.

This is not the 7nm GPU yous're looking for. We'll take more to say on compute specifically in the near future.

Now Read: