The RTX50 series has been released for a long time, with four models launched successively. However, shortages and price increases are still common, which makes many players feel tired and never want to play again. In fact, it is normal for new products to be out of stock when they are first launched. After all, production capacity needs to ramp up. However, the RTX50 series is particularly serious this time. Even the delayed release of the RTX5070 series has not alleviated it. Why is this?

According to analysis by industry insiders, there is not a single reason behind the scenes, but at least three different factors.

One is the RTX50 chip design flaw.

It is reported that,Blackwell GPU chips found to have warping and system failure issues, simply put, the top of the chip is not smooth enough, and naturally cannot be in close contact with the top cover of the radiator, resulting in heat accumulation.

This issue affects all Blackwell chips, not just RTX50 series gaming cards.Also includes AI accelerator card.

to this end,NVIDIA had to redesign the metal layer and bumps on the top of the GPU chip, and then re-tape and produce it., causing the mass production delivery of accelerator cards and game cards to be greatly delayed.

The second is the problem of ROP unit loss.

This problem is encountered for the first time in the history of graphics cards.

The first one to be found to have problems was RTX5090(D), and then it was discovered that RTX5080 and RTX5070Ti were no exception. Only the latest RTX5070 survived.

Loss of ROP units will result in a certain degree of performance loss, up to 11%.

NVIDIA acknowledged this problem, but emphasized that the proportion was very small, only 0.5%, but did not announce the specific reason.I don’t know if it’s a chip-level defect or if the quality inspection is too careless.

Regardless, a large number of graphics cards and chips were recalled, severely affecting deliveries.

The third is high-cost GDDR7 video memory.

The RTX50 series uses GDDR7 video memory for the first time, and the high cost is inevitable. However, NVIDIA has passed almost all of this cost increase to AIC manufacturers, resulting in almost no profit margins, and MSRP original price cards cannot be sold at all.

In fact, the demand for the RTX50 series is still very strong, but each brand has insufficient supply. The various shortages are not hunger marketing, but a helpless reality.

NVIDIA has informed various AIC manufacturers that the shortage is only temporary and the delay in delivery is only temporary. After the above problems are resolved, sufficient supply will be available, and it is expected to be significantly alleviated by the end of this month.

I hope that the subsequent releases of RTX5060Ti, RTX5060, and RTX5050 will stop making such a fuss.