Del's Unhinged Ranting

Just to be clear, this was an attempt to push voltage as low as possible with absolutely no regard for performance. I just wanted to see how low I could go while still technically running above stock clocks. I tried Steel Nomad and Time Spy for synthetic benchmarks, and a couple of games I frequently play for game tests. DLSS/FSR/XeSS were always OFF, all these tests are at native resolution.

-- IMPORTANT CAVEATS --

This is NOT a well controlled environment, this is me firing up games off my library, letting the temperature and framerate stabilise in a location that performs worse than average, then taking screenshots.

I do NOT have a good performance logging solution configured, nor do I have preset benchmark runs for these tests. They are quick and dirty and should be taken with more than one grain of salt.

The only thing that changed in between test runs was applying my custom profile on MSI Afterburner and returning to stock settings.

I expect the margin of error to be large, well above the 1-2% we usually see in professional benchmarks and reviews. I cannot verify exactly how much it is, so assume the worst.

Whenever "my overclocked profile", "overclocked" or anything with the same meaning is mentioned, it means the following settings in MSI Afterburner:

-- SYNTHETIC BENCHMARKS --

All tests run at default settings.

In 3DMark Steel Nomad, the card peaks at around 110 W for a decrease of 37% from stock, while performance drops by 13%. Compared to my overclocked profile, power draw went down by 41%, and performance went down by 18%. Results links: Undervolted (110 W), Stock (175 W), Overclocked (185 W). Note that the Undervolted run still has the VRAM overclocked for this test.

3DMark Time Spy had similar results. Performance loss was 14% vs stock and 17% vs overclocked, with power draw reduced by 43% and 46% respectively. Results links: Undervolted (100 W), Stock (175 W), Overclocked (185 W). Note that the Undervolted run still has the VRAM overclocked for this test.

-- GAME BENCHMARKS --

All settings and spots where screenshots were taken can be found here.

Assassin's Creed Mirage, custom Medium settings. The undervolted profile consistently draws 85 W or less even in more crowded areas. At stock settings the card draws between 175-180 W. That being said, this does come at the cost of significantly less stable frametimes, meaning this undervolt is probably on the absolute edge of stability and a bit more voltage would be necessary for a long-term setting. Note the over 20% performance loss alongside the nearly 500 MHz frequency drop.

Redout 2, max settings. Stock does incredibly well here, delivering over 100 FPS in most occasions. Tested at Neo-Tokyo, which tends to be one of the more performance-heavy tracks in my experience. Undervolting here dropped performance by up to 20%, and reduced power draw by 42%. Note that the overall frametime consistency has worsened here too, but it's fast enough that it doesn't detract from the experience.

Cyberpunk 2077, custom Medium settings. Stock settles in at around 60 FPS with my settings, while undervolting drags down to 50. Power draw drops below 100 W as a result.

-- HOW IT'S DONE --

Since Nvidia doesn't let you directly control the voltage of your GPU, you have to find a different way around it, and there is one:

Setting a maximum clock speed using MSI Afterburner's Curve Editor means your GPU will aim for that clock speed at the lowest voltage that matches it.

What this means is, if you make your V/F curve completely flat from one voltage onwards, your GPU will never try to go above that voltage. Thus, you effectively cap your card's voltage and select a target frequency for it.

For a visual example, here's my card's stock V/F curve:

And here's the undervolt setting:

There's a quick way to do this, so you don't have to set each little point manually. Select the leftmost point (lowest voltage) you want to try, set it to the frequency you want. Then, hold Shift, and drag across all the other points to select them together. Finally, hit Shift+Enter twice, and Afterburner will flatten the rest of the curve for you. Hit Apply and hope you don't crash.

-- THOUGHTS OVERALL --

Averaging out my results, power draw went down by 44% vs stock settings and 47% versus my overclocked profile. At the same time, the performance change was -18% versus stock. I didn't keep record of my overclocked gaming results, but the performance bump seemed to be around the same as that in synthetic benchmarks, if not a little less.

This makes for an efficiency improvement of 24% overall versus stock, accounting for performance loss versus power draw reduction. Not fucking bad, though again this DID really mess with frametime stability, so be mindful of that when trying this yourself. Frametime stability is just as (if not more) important than average framerates - A game running at 100 FPS with frequent stutters will usually not feel as good as a game running a rock-solid 75 or even 60, depending on the severity of said stutters.

Here are some of the promo images I was sent, and a few words about each. Forgive the heavy compression, I did get these over email.

The tuck cases. Oh man, the tuck cases. Of course they do the usual connecting thing that Virts tucks do, but this time the SS25 Spring and Summer tuck cases connect with each other as well as with themselves, forming different designs each time. Check it out:

I absolutely love the idea and the fact that they managed to pull this off. The Spring deck looks like the tuck case from the SS15 except starting to break open, and the Summer case looks like it has already blown up.

All this yapping and I haven't even shown you the cards themselves. Here you go!

SS25 Summer on the left, SS25 Spring on the right. The Summer deck is reminiscent of the P1/P1 Foundations/P1FX in its style, which I'm pretty happy with. The partial border was kind of a hit for me when I first saw it.

The Spring deck looks a LOT like the SS15, just with slight adjustments, the red and blue swapped, and a black internal ring to replace the previous red one:

It's basically a remaster, and in a way as much as I'm a little sad that I got into cardistry after the SS15 was already out of production, I'm glad I'll be getting my hands on this new 2025 Spring variant instead. It's not at all an original design, but it's a decent homage.

Finally, the face designs - These are taken directly from the P1 and P1 Foundations style for both decks, with the Summer deck not being numbered at all:

My opinion? I dig the Summer design. I love non-numbered faces, on Virts decks or otherwise. I love my Virts P1FX and Sea Glass, I love my Touch Origin, I love my Bicycle NEON, and this looks just as pretty to me. It does make the deck much harder to play with, but for a cardistry-oriented deck, that's a sacrifice I'm willing to make. Hell, the aforementioned Bicycle NEON and Touch Origin decks are completely impossible to play with as all the faces are the same, and they look absolutely gorgeous.

In short, the 9070 XT is maybe a tiny bit slower than the 5070 Ti, but also $150 cheaper.

In long:

(Image credits in order: Gamers Nexus, Hardware Unboxed)

It's within *earshot* of the RTX 5070 Ti for raster benchmarks, which is more or less what I was hoping for. With undervolting, Der8auer even managed to squeeze past the 5080 in 3DMark Time Spy Extreme:

Not bad. I don't expect the 9070 XT to win against the 5080 in any real-world benchmark (and it doesn't), but for $150 under the 5070 Ti? Not fucking bad.

There is one segment, however, where Radeon still gets absolutely pummeled: Ray tracing. The more RT heavy a game is, the worse the 9070 XT will perform.

It's immensely better than last generation was in this situation, don't get me wrong, but don't expect Black Myth: Wukong or Cyberpunk with Path Tracing turned on to run well.

(Image credit: Gamers Nexus)

With a less RT-heavy workload, the 9070 XT finds its footing again, but still doesn't beat the 5070 Ti:

(Image credit: Gamers Nexus)

The new media encoder also seems to be doing well, with image quality in H.264 not quite where the competition is but really close:

Meanwhile, in AV1, it's almost the exact same:

Considering that Twitch lets you stream at 6000 Kbps (and they won't stop you if you set your OBS to 8000), while YouTube lets you hit 10000 Kbps, this encoder is completely acceptable.

Overall? I love it, it's more or less three times as fast as my RTX 2060 Super and it has double the VRAM, which means I should be able to stream Redout 2 and Ghostwire: Tokyo without Unreal Engine being a bitch about it.

Pricing the 9070 XT at $600 is acceptable, and really nicely undercuts the 5070 Ti. Considering rumours show the 9070 XT being just a tiny cut under the 4080 Super (and thus 5080) on average, at $150 cheaper it's gonna fight well even if it isn't the "shut up and take my money" that it would've been for $500.

For the 9070 on the other hand... They didn't try at all. It'll just have to be faster than the 5070 since it's launching at the same price, but if it's not, it'll be a miss. This card would've absolutely murdered at $400 and it's a shame seeing AMD just forgetting to undercut the competition at all.

Nvidia has had the absolute worst launch possible. Cards failing on driver installation, cable current imbalance leading to melting connectors that's unfixable without a hardware revision, absolutely no stock, the entire 50 series stack occasionally missing physical hardware, prices not being real at all and partners scalping their own cards, it's an absolute joke. AMD had an opportunity here to come out swinging and storm the market by undercutting both their lower offerings by $150 flat and steal MASSIVE market share, but they chose not to.

It won't be a massive swing over, but at least it'll move more people away from the growing monopoly that is Nvidia.