xt Yingtong Swimsuit Unboxing: Real User Experience & Detailed Review

I’ve owned this graphics card for about a month and a half now, and after all my tweaking and testing, I’d say it’s safe to conclude that I’m thoroughly satisfied. At first, I was torn between the 5070ti and 9070xt, even considering upgrading to the 5080. Eventually, I spotted the card in stock—albeit not at its original price—but as the last unit available, I couldn’t resist the allure.

I decided to spend a little extra to secure it sooner rather than later. Initially, I paired it with an older motherboard and CPU, but something felt off during installation. So, I ended up replacing everything.

When I first unboxed it, there was a noticeable perfume-like scent. After running various benchmarks, I found that the performance metrics were slightly lower than what I’d seen in online reviews, scoring around 28,000 on 3DMark. I began to suspect that my 9900k might be bottlenecking the system. Switching to the 9950x improved the score marginally.

With all default settings and the OC button enabled, the power consumption settled at 340W, pushing the score to 29,700. Perhaps my card has a slightly weaker build quality. In a sealed room resembling a seaside apartment, the core temperature hovered around 57°C under default stress tests, while the VRAM temperature climbed to 86°C.

Initially, I envied setups with vapor chamber cooling, but realized my core wasn’t overheating—it was the VRAM. I wondered just how much improvement a vapor chamber would bring.

Next, I ventured into overclocking, experimenting under negative voltage and various adjustments. I managed to achieve scores of 33,000 in benchmarks, but experienced driver crashes during gaming sessions, with no significant frame rate improvements. Eventually, I reverted to default settings, fine-tuning some parameters conservatively until I stabilized at over 30,000 in benchmarks.

Even after a week of gameplay, occasional driver crashes persisted, leading me to suspect my power supply. My original Corsair RM850x, a trusty 850W unit that had served me for six years, seemed like it might be nearing its limits. Deciding to replace it altogether, I also upgraded my case and installed a 1200W PSU.

Since then, I haven’t experienced a single crash.

Returning to overclocking experiments, I discovered that my card would crash during gaming if the negative voltage exceeded 50, yet it handled 3DMark tests flawlessly even at over 100 negative voltage. Stress tests passed without issue, but since gaming instability remained unacceptable, I dialed back to extremely conservative parameters.

To date, I haven’t encountered a single crash. During gameplay, the VRAM temperature typically ranges between 85-86°C, peaking at 88°C. Lowering the VRAM temperature could involve increasing fan speeds or reducing power, but I opted to simply ignore it—out of sight, out of mind.

There’s a common belief that A-series cards struggle with online games, but in my personal experience with CS2, I noticed no such issues. It could be the CPU picking up the slack. Running CS2 benchmarks at 1080p yielded an average of 794fps (with a negatively voltaged CPU achieving a R23 score of 45,000), fluctuating between 400-550fps in the town map.

I’m quite pleased with these results. Contrary to online claims, I haven’t encountered any frame drops while playing CS2 regularly—if I had, I’d have switched long ago. Playing 4K AAA titles presents no major hurdles either. For example, in “Wilderness,” running full settings including ray tracing, FSR4 Balanced + Frame Generation + AFSS nets me around 180fps, while turning off all enhancements leaves me at about 50fps in pure raster mode.

In summary, this generation of 90-series A-cards remains impressively competitive (especially when comparing raster performance against competitors).

Ultimately, purchasing advice depends on individual needs. If you’re considering a switch, keep an eye on the upcoming 60-series N-cards before making your decision.