Zotac RTX 5080 Solid Overclocking, Thermals, Noise, & Gaming vs. Founders Edition (Review)
jimmy_thang
February 17, 2025
We perform a tear down on Zotac’s RTX 5080 Solid video card and analyze its thermals, frequency, acoustics, gaming performance, overclocking, and more
The Highlights
- Zotac’s RTX 5080 is noticeably larger than the Founder’s Edition model, taking up over 3 slots
- The card costs roughly $270-$300 more than the Founder’s Edition model
- Zotac’s RTX 5080 Solid generally performs about 3-6% better than the 5080 FE
- Original MSRP: $1,270 (approximately)
- Release Date: January 30, 2025
Table of Contents
Visit our
Patreon page to contribute a few dollars toward this website's operation (or consider a
direct donation or buying something from our
GN Store!) Additionally, when you purchase through links to retailers on our site, we may earn a small affiliate commission.
Intro
Today we’re reviewing the Zotac RTX 5080 Solid. You should not buy it. At the time we wrote this, it’s being sold for $1,270 on Newegg, which is a ripoff for an RTX 5080. That’s 27% more than MSRP, and the 5080 was already a hard sell at $1,000. A 27% hike is horrendous value, and that’s about the start and end of our review. Thermally, the card is fine. Acoustically, it’s also fine. But the price is a ripoff.
Zotac has two main models of 5080: The Extreme Infinity and the Solid. We’re reviewing the Solid OC model, which has a boosted clock. The marketing is amusing for both the Solid and the Extreme Infinity
Editor's note: This was originally published on February 11, 2025 as a video. This content has been adapted to written format for this article and is unchanged from the original publication.
Credits
Test Lead, Host, Writing
Steve Burke
Testing, Camera, Editing
Mike Gaglione
Camera, Video Editing
Tim Phetdara
Writing, Web Editing
Jimmy Thang
“To gaming and beyond!” “Infinity Mirror,” “Spectra ARGB,” “Icestorm 3.0,” and “up to 9 composite heatpipes,” because that sounds better than explaining that the 5090 and 5080 are different. It even has a “card-length heatsink,” which is good because we’d rather have that than a backplate and fans extending past the heatsink... And yes, the PCB is shorter than the heatsink.
Despite its price, we’re still going to test this card and see if it’s designed well, we’re going to tear it down and look for problems. We’ll test it thermally and acoustically as well.
We’re borrowing the card from a system integrator because we don’t want to work with the partners directly for them these days and we haven’t been able to grab one when they go on sale. Supply appears to still be about 0. Let’s get into the testing of the Zotac 5080 Solid.
Overview
The last few Zotac cards we looked at had major oversights of design. The worst one had a huge heatsink that didn’t contact the MOSFETs. This card didn't have that issue.
Zotac’s website says the “Solid” series is “no frills, all action.” We don’t know what that means. It’s a video card.
The card also has “a powerful cooling system” “*available on select models.” Good. We like the uncertainty of the product page specific to this product.
The site also says:
“With a stealthier aesthetic and engineered for durability and performance, the ZOTAC GAMING SOLID is for those who demand pure performance over anything else.”
As for actual features, it has an LED that alerts users of an improperly inserted 12V-2x6 connector, a dual VBIOS switch, a partial flow-through area on the back, and uses a massive heat sink. Unfortunately, the card uses a lot of plastic, including on the back and its design blocks some of the exhaust.
Zotac RTX 5080 Solid Tear-Down
Zotac’s RTX 5080 Solid is much larger than the 5080 Founder’s Edition version (read our review), taking up over 3 slots. One nice thing about the card is that it uses a larger bracket, which helps with structural rigidity.
The card uses a typical 3-fan design but has a somewhat large flow-through area on the back. Unfortunately, Zotac’s design blocks off roughly 20% of the card’s flow through here.
The card comes with a 3 8-pin adapter for the card’s 12V-2x6 connector.
The card has a unique plastic “grill.”
The back is similar with some perforated holes, but underneath, you’ll see a lot of plastic obstructing airflow.
Measuring the fans, they are 95mm ones and have a ring around the outside, which are supposed to help direct the airflow straight through a little more. The first 2 fans are going to push air straight into the PCB in a more traditional way with the third fan pushing some of the air out the back.
To remove the aluminum backplate, we simply remove several screws on the back, which expose approximately 2.5mm thick thermal pads that contact the backside of the memory and VRM.
Next, we removed the card’s backplate by removing some screws.
To remove the front plate, we, again, removed some screws.
Moving forward, we removed one cable on the back.
From there, we removed the GPU’s leaf spring.
Removing this section of the PCB exposes 2 connected fan connectors, which we need to disconnect to proceed further with the tear down.
From here, we can see plastic around the GPU die, which is uncommon. We can also see that the card also has 2 additional 2mm clay-like thermal pads that cover the card’s inductors.
The card also has 3 Zotac-branded fuses. From what we’ve heard a while back, these fuses won’t act fast enough to protect the card but does help the RMA center help identify where the failure happened.
Flipping the card over, we can see a vapor chamber. We also see the memory contacting a separate plate.
The GPU uses seven 6mm heat pipes, which is a lot.
The card has thermal pads that make contact with the MOSFETs, which Zotac had an issue with in the past.
We also noticed a thermal pad that touches the card’s R10 inductor.
The card’s fan cables are very accessible, which is good in case you have a fan die at some point.
Cleaning off the thermal paste from the 5080 GPU, we can see the GB203-400-A1 die, which is a lot smaller than the 5090 die.
Next, let’s move into our thermal testing.
Zotac RTX 5080 Solid Tests
Grab a
GN15 Large Anti-Static Modmat to celebrate our 15th Anniversary and for a high-quality PC building work surface. The Modmat features useful PC building diagrams and is anti-static conductive. Purchases directly fund our work! (or consider a
direct donation or a
Patreon contribution!)
Thermals - Equilibrium
We ran this prior to the tear-down in order to preserve the card as-is for testing. We tested both VBIOS options for thermals and frequency.
In a quick head-to-head, tested using the auto settings and whatever VBIOS tells the card’s fans and clocks to do, we landed on these results. The 5080 Founders Edition card ran at 32 dBA in our hemi-anechoic chamber and pulled 315W board power under this workload. Those variables put the GPU temperature at 65 degrees Celsius and memory at 72.8 degrees Celsius. The Zotac 5080 Solid tested with the same software and auto conditions ran at 31.1 dBA and 360W. The workload doesn’t pin these to TDP, but is consistent.
The Zotac card runs a higher out-of-box power budget and its higher clocks are pulling more power. Even with the higher power draw and marginally lower noise levels, the Zotac 5080 card manages lower temperatures than the 5080 FE. Considering the enormity of the Zotac card, this is what we need to see. The 4-degree reduction in GPU and 7-8 degree reduction in memory thermals gives the Zotac Solid an advantage, although not in size, but the size is working in its favor.
Frequency
This chart shows the frequency over the test pass. The FE card plotted around 2650 to 2680 MHz on average, holding relatively stable under this test workload.
For reference, the 5090 has a lower and spikier clock. The configuration is much larger and the 5090 makes up for the difference in other ways.
Adding the Zotac 5080 solid to the chart, its clock is higher than the NVIDIA FE model, landing at 2790 to 2842 MHz. This is a pretty significant uplift. We’ll test that in some performance scenarios momentarily to see what it actually does.
Noise Frequency Spectrum
The next chart is produced using our hemi-anechoic chamber that we built with support from our Patreon backers and our supporters on store.gamersnexus.net, which allows us to isolate noise and ensure a like-for-like testing environment day-to-day.
This chart shows the 5090 Founders Edition and the Zotac 5080 Solid. The 5080 Founders Edition is mostly equivalent to the 5090 Founders Edition, with the exception that the fan curve runs slightly slower and so the noise is reduced by about 1 dBA in our workloads.
The Zotac Solid doesn’t encounter the same 180 Hz spike we saw in the 5090 and 5080 FE coolers, but does have its own spike around 433 Hz. The curve generally follows the same path, with marginally higher noise levels in the 1000 Hz to 1600 Hz range, but lower noise levels in most other places. The Zotac Solid overall manages to keep its noise levels down and avoids being obnoxious. It is making use of its size.
VBIOS Comparison - Frequency
This is a quick VBIOS comparison for the Zotac 5080 Solid. Between the two VBIOS options, frequency is almost completely identical. There is no meaningful difference.
VBIOS Comparison - Thermals
For thermals, the silent VBIOS runs about 3-4 degrees warmer on the GPU from its reduced fan speeds. The fans run about 4-6 percentage points slower than the default VBIOS. The memory temperature climbs about 3 degrees. There’s not a big impact to thermals from the VBIOS change and the fans don’t really change a lot either.
Zotac RTX 5080 Solid Gaming Benchmarks
Resident Evil 4 - 4K
In Resident Evil 4 at 4K, the RTX 5080 Solid ran at 128 FPS AVG and led the 5080 FE by 4.2%. That’s not a bad uplift, but it’s bad for almost $300 more, though. Lows are within usual variance and benefitted similarly.
Resident Evil 4 - 1440p
At 1440p, the 5080 Solid held 237 FPS AVG against the 5080 FE’s 224, which is a 6% lead. That’s a big jump within the same GPU model. Again, the price isn’t justified, but at least we’re seeing an impact from the pre-overclock by Zotac.
Black Myth: Wukong - 4K RT
In Black Myth: Wukong at 4K and ray traced with upscaling, we saw the RTX 5080 Solid at 60 FPS to the FE’s 58.6. That’s a 2.6% improvement over the FE model.
Black Myth: Wukong - 1080p RT
At 1080p and with RT with upscaling, the Solid ran at 128 FPS AVG to the FE’s 122. That’s a 5.5% improvement over the FE model.
Final Fantasy XIV - 4K
In Final Fantasy 14 Dawntrail at 4K, we saw the 5080 Solid at 117.5 FPS to the FE’s 112.4, a 4.5% uplift. This trend repeated across the other games we tested. It is consistent, so we’ll stop here with the game benchmarks as it tells the same story. In short, it’s generally about 3-6% better.
Overclocking - Steel Nomad
Overclocking has some bugs on the 50-series right now. We haven’t dug too much into this bug, but it appears as if the voltage offset slider has an issue with the 50-series at present and reduces GPU clocks when used. We ended up setting it to 0 instead.
This quick table shows the overclock stepping with 3D Mark Steel Nomad. From stock, we were able to hold an offset of 700 MHz core, which is huge, without a memory offset, landing the result at 8160 points. The uplift from stock was an insane 28%. This was not replicated in real-world use cases so check your expectations. Adding memory clocks introduced instability without dropping the core offset to 650 so we did that.
Unfortunately, we weren’t able to use the 700 MHz offset with the memory offset. We also were unable to find stability from these clocks in any gaming use case, so sadly, it was purely synthetic. This is mostly interesting from a competitive overclocking standpoint.
Overclocking - FFXIV 4K
The next testing was only done for a performance check. We did not burn this in for stability, so we want to really heavily emphasize that this overclock is not stable in all games we test. It was only stable in Final Fantasy XIV: Dawntrail at 4K. You’d have to spend time tuning it down to dial it in, but we wanted to present what we found so far.
The stock 5080 FE ran at 112 FPS AVG. The 5080 Solid, out of the box, ran at 117.5 FPS without any background applications. We saw performance degradation when we began running Zotac’s utility in the background in addition to changing the clocks and fan speeds, so just increasing the power offset and applying up to a 200 MHz core offset resulted in performance equal to or worse than stock. We were able to run the Final Fantasy benchmark with stability up to a 500 MHz core offset and 1200 memory offset; however, we’re not certain if there are any invisible background issues with this memory offset especially out of this game.
The end result was a 125.7 FPS AVG, which led the stock Zotac Solid result of 117.5 FPS AVG by 7%. That’s not a bad uplift but it’s not that exciting. The lead over the FE model is almost 12%, which is more meaningful.
Overall, that’s a decent amount of OC headroom from the base FE result, while the 7% uplift is typical for recent generations. In no real-world case were we able to replicate the Steel Nomad uplift.
Overclock Frequency Results
Finally, here’s the frequency chart in Final Fantasy. This isn’t comparable to the earlier charts for frequency. Out of the box, the card ran at 2625 MHz. Memory ran at 1875 MHz, but keep in mind that there’s a multiplying effect for GDDR.
The core after a 500 MHz offset ran at 3120 MHz, which is almost 1-to-1. The memory after offset ran at 2025 MHz. This is about correct and what we’d expect after multiplying the numbers for GDDR effective speeds.
As an important note here, this type of testing mostly looks at the impact from the power percent slider offset. Broadly speaking, GPU variance from the “silicon lottery” has more total impact than any particular partner model outside of its power boosting headroom (and the super high-end models sometimes being binned, but this is rare on anything but flagships).
Zotac RTX 5080 Solid Conclusion
Grab a
GN Soldering & Project Mat for a high-quality work surface with extreme heat resistance. These purchases directly fund our operation, including our build-out of the hemi-anechoic chamber for our acoustic testing! (or consider a
direct donation or a
Patreon contribution!)
The Zotac RTX 5080 Solid is built well overall. We didn’t see any massive red flags with this card except for its price. If you don’t care about that and you just want a 5080, this card is fine, again, except for its price. We didn’t have any major complaints with its thermals, assembly, or acoustics.
The thermals, in particular, are a little bit better than the Founder’s Edition card, but not much better as they probably should be given the size of the card. The memory was improved a decent amount, however, being roughly 7 to 9 degrees cooler.
Performance is roughly 2-5% better out of the box than the Founder’s Edition card.
In the past, we found Zotac’s MOSFETs weren’t contacting the thermal pads for the heatsink, but that, fortunately, wasn’t the case here.
Overall, there was nothing hugely detrimental here aside from the value as $1,000 for a 5080 is already very questionable. It’s hard for us to compare the thermal performance when we only have this partner model. We haven’t been able to get any others yet, but that’s something we’d like to dive deeper into provided we can get more cards in.
Connexion
