I would like to slightly correct the previous answer about water resistance. The device was first designed and has been advertised to have at least IP67+ water resistance.
During the enclosure design refinements we have decided to fill the device entirely with epoxy compound. Our aluminum enclosure already has thickness ranging from 2.0mm to 6.5mm and this further prevents enclosure deformation - effectively moving AsicVault device into diving watch territory. However, we currently don't have immediate plans to test our hardware wallet under water at deep diving depths.

Of course it is legit, I stand behind every single claim made on our website https://asicvault.io

How can it be 1,000 times harder to crack compared to other BIP-39 hardware wallets?
BIP-39 hardware wallets are working on very low performance microcontrollers or secure elements.
They are doing only 2,048 iterations of PBKDF2 SHA-512 that is even less than old NIST recommendation of 10,000 rounds from year 2016.
Performing higher number of PBKDF2 SHA-512 is standard practice for good security. iTunes does it, LastPass does it and Veracrypt as well.
Even Ledger agrees that this very low number is the main problem of BIP-39:
https://donjon.ledger.com/Unfixable-Key-Extraction-Attack-on-Trezor/

AsicVault specially designed SHA-512 accelerator inside high performance secure chip is at least 340 times faster than common microcontrollers.
The number of PBKDF2 SHA-512 rounds is set to be exactly 1,000 times higher than BIP-39, hence the cost to crack AsicVault is also 1,000 times bigger.
Please read in-depth teardown review and validation of AsicVault SHA-512 performance here:
https://medium.com/@Zero404Cool/teardown-of-two-touchscreen-hardware-wallets-trezor-vs-asicvault-a92923a5b6ce

If it takes 10 days to crack an ordinary hardware wallet, it would take 10,000 days ie 27 years to crack AsicVault.
If it costs $10,000 to crack an ordinary hardware wallet, it would cost $10,000,000 to crack AsicVault.
You can perform independent analysis according to the PDF document in video description and our device performance is shown on this video below:
https://vimeo.com/293211785

What is SRAM-PUF and why do we need it?
While some people have said that there is no point of encrypting the seed phrase and it is totally fine to hold it in RAM, we think the opposite.
The main reason given for not encrypting seed phases has been that the encryption key has to be stored in Flash and it is as easy to extract.
This is true for non-secure microcontrollers that lack the sophisticated hardware required for proper key storage, like SRAM-PUF.
Physically Uncloneable Function does not store the encryption keys in non-volatile memory, they are invisible when the device is powered down.
https://en.wikipedia.org/wiki/Physical_unclonable_function
This way of deriving a key from the SRAM properties has great security advantages compared to traditional key storage in non-volatile memory.
Because the key is not permanently stored, it is not present when the device is not active (no key at rest), and hence cannot be found by an attacker who opens up the device and compromises the memory contents.

AsicVault does not rely just on PUF, we are using multiple layers of encryption everywhere.
We have a very fine unique 75nm mesh around the electronic components protecting them actively from outside intrusions.
If you destroy the mesh, your encryption key will be broken as well.

AsicVault can be called a true hardware wallet since we have designed all key components in hardware, not in software.
We have written over 77,000 lines of Verilog/VHDL code (hardware design), together with testbenches over 550,000 lines.
This is more than Bitcoin Core client code size - all in hardware.

Thank you eternalgloom, I personally like the design a lot too.  Here are some recent pictures:





So, challenge me about anything on the technical side and I am ready to prove it, explain it in detail and show it to you.