Hey guys,

Just wanted to gauge interest in a new RAM OC tool I’ve been cooking up: FROST (FranceOC Resource Overclocking SPD Tool).

Since Thaiphoon Burner has been pretty much abandoned, I decided to build a proper alternative from scratch with a modular approach. The idea is to use hardware modules for different memory types and I2C/SMBus hardware, making it as versatile as possible—from old-school SDRAM to DDR5, GDDR6, and hopefully beyond.

It relies on a dedicated hardware dongle plugged into the DIMM/hardware, paired with custom software.

What it will do:

• Read, dump & flash full SPDs

• Accurate IC detection (Samsung B-Die, Hynix AFR/CJR, Micron E-Die, etc.)

• Edit JEDEC timings & tweak XMP profiles

Quick output preview:

• Module type: UDIMM | Ranks: 1

• Bus width: 64-bit | Component width: 8-bit

• Capacity: 8.0 GB | Estimated speed: DDR4-3200

• Timings (ns): tCKAVGmin: 0.625 | tAAmin: 10.0 | CAS: CL16

Got a first working prototype on the bench right now. I'll drop more details and testing logs as soon as I wrap up the initial hardware checks, aiming for a v1.0 release shortly.

What do you guys think? Any specific features you'd love to see in it?

Edited August 5Aug 5 by IMtech

I don't really understand the problem you are aiming to solve with this product. Could you give more explanations ?

1 hour ago, Paul7347 said:

I don't really understand the problem you are aiming to solve with this product. Could you give more explanations ?

Good question. The goal is to provide a low-cost hardware tool for reading, backing up and recovering SPD EEPROMs directly from memory modules, without relying on motherboard support.

Software like Thaiphoon Burner works well when the RAM is still accessible by the system. However, if the SPD is corrupted, write-protected, or the module is no longer detected by the motherboard, software-only solutions become much less useful.

This project uses a small RP2040-based USB device to communicate directly with the SPD EEPROM over I²C. It can be used to:

• Read and back up SPD contents

• Analyze JEDEC/XMP data

• Verify SPD integrity

• Recover or reprogram damaged SPD data

• And much much more

In short, Thaiphoon Burner is mainly a software tool that depends on a compatible platform, while this project aims to be an inexpensive standalone hardware interface for SPD diagnostics and recovery

IMO, it's a pretty useful tool, but only for a specific group of people. For example, I've been tinkering with SPD for a while, turning some crappy DDR4-2133 modules into 3000-3200 MT/s sticks that work OOTB. For money, ofc.

How are you going to deal with SPD Write protection?

6 hours ago, ProfessorZoom751 said:

IMO, it's a pretty useful tool, but only for a specific group of people. For example, I've been tinkering with SPD for a while, turning some crappy DDR4-2133 modules into 3000-3200 MT/s sticks that work OOTB. For money, ofc.

How are you going to deal with SPD Write protection?

Your use case (retesting/rebranding memory modules) is exactly the kind of advanced application I aim to support eventually.

To be honest about the current state: FROST doesn't yet handle writing RSWP/PSWP protection settings, but it's on the roadmap.

Technically, there are two separate layers:
The PC chipset's "SPD Write Disable", which doesn't apply to us since we perform read/write operations out-of-system in this version of FROST. However, the RSWP/PSWP feature on the EE1004 type chip itself requires raising the A0 pin to ~9V during the command—something a small dedicated boost circuit will be needed on the future PCB revisions.

So: reading the protection status (RSWP status) will be easy to add to the existing protocol. Enabling or disabling the protection will require a hardware revision to include a high-voltage generator for the A0 pin. It’s on the roadmap but hasn't been implemented yet—if you already have a working circuit for this (voltage, timing), I’d love your feedback; real-world usage beats my datasheet research any day of the week

Edited August 7Aug 7 by IMtech

On 8/7/2026 at 10:54 AM, IMtech said:

if you already have a working circuit for this (voltage, timing), I’d love your feedback

Unfortunately, I don't have "my own" way to bypass hardware write protection. I marked such modules as "rejected" precisely because they had a hardware write lock, which I didn't want to bypass - the profit was too small for me to want to do so. Those that had fully unlocked SPD (Kingston for example) I first overclocked, checked their stability and then modified their SPD dump so that they worked OOTB with the given settings.