NVIDIA’s RTX 50 memory clock limit has been bypassed by none other than Teclab, allowing their RTX 5070 Ti to hit 36 Gbps speeds.
The folks over at Teclab have been showcasing some awesome mods with the latest NVIDIA GPUs and have also broken various world records. The tech and overclocking enthusiasts from Brazil were the first to mod a GeForce 10 series card with higher VRAM, and also unlocked higher memory clocks through custom modding.
Now, Team Teclab is coming out with a brand new mod which aims to bypass the memory clock limit set by NVIDIA on its GeForce RTX 50 GPUs. The RTX 50 GPUs feature GDDR7 memory and feature 28 Gbps to 30 Gbps speeds. The effective clock speed is reported at 15000 MHz for the RTX 5080 and 14000 MHz for the rest of the RTX 50 cards.
Currently, all overclocking utilities allow a +3000 MHz overclock limit on the memory, which cannot be exceeded since it’s locked by NVIDIA. Previously, modders have replaced the memory chips with higher frequency modules to gain faster speeds. Other than that, Teclab themselves have modified the hex code to unlock higher limits. But this time, they have a new mod.
As Burti_TecLab explains, he used the most entry-level model offered by GALAX, the 5070 Ti 1-Click OC. The card isn’t shunt modded by; its power limit is completely unlocked. The software wasn’t able to read the power consumed by the card, as the multiplexer that reads the current was disabled. The first run was scored without any overclocking applied, but the power limit was unlocked as mentioned.
The second run was using a manual overclock, with the card boosting above 3.3 GHz. The third run is where the magic happens. How the bypass works is by tricking the clock management of the GPU on a logic level or the initial programming level. So with this mod, the GPU will think that it is running at a base or lower frequency when, in fact, it will be set to a much higher frequency. This applies to both the GPU and the memory clock.
Even though the monitoring utility was showing the GPU clocks lowered down to 3.1-3.2 GHz and the memory at 28 Gbps reference speeds, the actual score was even higher than the previous manual OC. It shows that the logic board was actually maintaining a higher clock speed, which is confirmed by Teclab themselves, who state that the RTX 5070 Ti was running at speeds beyond 36 Gbps (18,000 MHz+). The numbers breakdown in Unigine Superposition is as follows:
- RTX 5070 Ti (No OC/Power Limit Unlocked): 9922 Points
- RTX 5070 Ti (+500 GPU / + 3000 Memory /Power Limit Unlocked): 11722 Points
- RTX 5070 Ti (+330 GPU / 36 Gbps+ Memory / Power Limit Unlocked):11993 Points
It’s really awesome to see the Teclab team showing how it’s done with bypassing the limits set by NVIDIA for its RTX 50 GPUs. One should remember that 36 Gbps is just insane for GDDR7 since these are 28 Gbps modules. We know from reports that Samsung, Micron, and the rest are working on 36 Gbps memory that will feature higher VRAM densities, but aren’t expected to roll out till late this year or may even be pushed back to 2027 if the memory shortages continue.
Meanwhile, these 28 Gbps having the OC potential of up to 36 Gbps is simply amazing, and Teclab is going to present even higher OC capabilities and records in an upcoming video where they will be testing the RTX 5070 Ti HOF from GALAX. We look forward to their next OC venture.
Follow Wccftech on Google to get more of our news coverage in your feeds.
