adaseb
I am not too technical with the details
But that didn't stop you from calling the project a scam Wink

So miners end up hashing this code and if they find a block, the hash originally contains the pool info in the paramter.
ExtraNonce, yes, this is a problem in bitcoin's stratum1. But in ETH, the situation is somewhat different.
EthereumStratum1 does not have a mandatory ExtraNonce (except the mining.set_extranonce notice that can be rejected), but it have etherbase and task ID.
The etherbase hashed, but task ID and message id - not. Obviously, the solo pool will not accept a task with an incorrect ID and we have to repack the data asap.
It is expensive to have different nodes (for PPLNS and SOLO) and sometimes pools are tricky here. The daemon is the same, but the task changes a little. From the outside, it looks like different nodes, but a little traffic analysis allows you to determine this trick and use this. The task changes every 3 seconds (or less) and so I use Cyclone IV FPGA

This is the tip of iceberg known to everyone. There are also lesser-known vulnerabilities that I exploit. In any case, the ShareSwitch scans the pool first and tries to configure the most profitable exchange. And only then analyzes and changes the traffic. If the pool is well protected, this will inform the user.
Each pool's admin is lazy in his own way. And each pool has its own vulnerabilities. EthereumStratum2 has a range of ExtraNonces for each miner, for example. This opens up new possibilities and closes some outdated ones. Some of these vulnerabilities could have been closed years ago, but the developers are too lazy... Most likely, it will never be closed again, because there is a POS ahead.

If you don't believe me choose some low hashrate ETH clone type coin OR fork your own and see if you can actually succeed in doing this.
Wrong way. It does not depend on the hashrate. However, ETC has even more holes, eah...

Mine on your local pool and if you find a block, try and send that block to the network with your solo client, and see what happens.
I've explained to you how it works. It's a little more complicated than you imagine

But for example in Bitcoin Stratum V2 such a scenario is possible after some manipulations. I have a video where ASIC works in this mode.

JayDDee
used for eth it would be so obvious and easy to exploit it would be common knowledge
It is not easy to use and it has been known for a long time. It's just that every teenager could do it before, now you need an FPGA for this.

The only reason to use an FPGA is to steal the block from the miner who tried to sleal it in a true MIM attack, ie attack both ends.
So, can I steal the block or not? Wink

This device is placed between the miner and the router. There are thousands ways to uncover unnecessary connections and traffic. And of course it can be blocked at the router level. What is the point of such a scenario?

That's *if* such a bug exists, and if it did it would have destroyed Eth mining by now.
I'll just remind you that CryptoNight ASICs (Baikal Giant-N etc.) existed long before the official release and everyone claimed that this impossible.
The first FPGA miners for bitcoin did not appear in the RadioShack.
This device has been around for a long time too. This only manifests itself if you have found a block. If not, no one will notice the changes. Don't worry, POS will kill mining much earlier Wink

malefactor
You don't like FPGA, I see

He copied some ideas/text from how real FPGAs are done.
The real FPGAs are done on the factory. Your question about bitstream maybe?

OK, this is because you need microsecond response times?
Because it contains a lot of operations that the AR9331 will not be able to handle. AEAD, for example.

Software-driven-implementation delays on the order of miliseconds don't matter for mining.
Yep. But without FPGA, it will take seconds. The task changes every 3 seconds or less in ETH, do you know?

You are selling a toaster, it does one thing only, and it's unique with zero competition.
The iPhone is also very simple. Plastic and glass only. If you analyze it without any knowledge, of course Wink

It's far easier to implement everything you describe in software
Again an easy implementation of an impossible technology?.. Please decide: either it's impossible, or it's easy to implement?

you can't make the argument for asking for pre-orders because you have to fund physical device development (which is of course very expensive, made more expensive due to component shortages), can you?
The development is completed in April, check the publication dates. It could have been ready in May, but I couldn't buy the components until July.
No pre-order has been paid. No one has paid before. Keep your fantasies to yourself.

By "endpoint" you mean "miner". But "endpoint" sounds more technical, because that's hardware lingo
By "endpoint" I mean endpoint. Learn the terminology first.

PCIe CEM spec among others
USB have endpoints and many other interfaces. It is a good practice to use Google before posting. No need to thank me, sir Roll Eyes

All you're claiming to do is mine on two pools at once. It's not rocket science. It's not even particularly hard.
For all the mining platforms/OS in one box? You can implement it at any time and compete with me!

why do you need microsecond speed on a task that can see milliseconds of delay without issue?
Because I know a little more than you do.

Bottom line: usec speed isn't a requirement to do the task. The use of an FPGA provides no value and is a complicated, difficult solution to something that can be easily done in software.
I will look at your developments with undisguised pleasure. Especially in FPGAs.

sergioheadache
id like to preorder
The devices are in production now. The sale is open on our website, welcome.

embedder

Posts

Merits