Tried latest Win version (_x64). Activated .NET etc.
Application terminates immediately without any sign if error messages, log messages or anything.

Please add lyra2z / GENTARIUM

https://gtmcoin.io/

It's a top contender for lyra2z profitability at the moment.

I have been following this project since it was announced.
I would recommend you try and update the first page of this thread and also the web page.
The whitepaper is also totally confusing. It's a mess with screenshots, user guides for various things, ICO info, presentation bullets and all sorts of incoherent information.

It's very difficult to understand the current status of this project. Maybe do like all other projects and make a simple roadmap and check the boxes as they are completed.
If you have software that can be downloaded and executed, put it on github and document it there.
Use the whitepaper to explain the architecture

Restored to blockchain drivers.
Managed to get pp tables and everything working in short order to restore things as they were. So, I haven't lost my touch yet. It's AMD who sucks... lol

Switching to ETH for now.
Enjoy my absence of hash power !

Pffft
Just as anticipated. I upgraded from blockchain driver (or should I say downgraded) to 18.5.1

Hashrate went down compared to before and power consumption went up by 60%. Powerplay settings or voltage control obviously has no effect.
My server PSUs started to sound like F16:s spooling up and the Vegas ran the fans at full speed to keep up with this crap.

Every time, I reboot, the system behaves differently. Device ID:s not recognized, Cast locks up immediately, shares not accepted, the entire system typically crashes after 3 minutes and auto reboots.

Seems this driver is completely unstable. Tried switching to mining ETH instead. Computer crashes during DAG file creation.
AMD never ceases to produce useless software.

Anyone want to buy a bunch of Vegas?

Just ran into the problems with hashrate/V8/blockchain driver

So we need to finally update drivers. For a Vega owner, that may result in endless pain and lifelong suffering but ok...

Do the new drivers work with powerplay tables and other tricks we are used to?

I hope to be able to start verifying the performance difference between 16x and 8x on real AI workloads within a few weeks.
I'm also looking for boards with lots of well spaced slots with full bandwidth and I'm not alone.

I suggest Threadripper and PCI switches (e.g. PLX) like what was done on some X99 boards in the past.
Other important factors for AI use cases is provisioning for large amount of RAM. Typically, you want 2x the amount of memory on the GPU.
Storage is another important parameter so bandwidth has to be reserved for that. If the compute models can't be fed data from storage as fast as it's being processed, all is for nothing.

The business case for distributed compute is not a big mystery.
If this team were to try and do the compute themselves, they would have to compete with existing traditional datacenters. Since SONM have no technical advantage over a traditional datacenter, they would not be a able to compete if they built their own dc.

Look at the similar project DBC (DeepBrainChain), which is developing ditributed AI mining. They mix this up with an investment token (DBC) and a big ICO. Probably just to secure their Lambos in case the distributed compute doesn't work out. Recently, they got this brilliant idea to make even more money by bypassing the distributed concept and play on the greed of miners/investors. They are selling pre-built, very high end AI hardware and hosting services. They give priority to people with big DBC holdings. Basically, they are building a high end traditional datacenter with other peoples money. No economical or technical risks and they can start computing before they solve the distributed compute problems. They have sold thousands of AI compute machines this way. (Way higher spec than 1080 Ti).
Chinese Entrepreneurship at it's best and maybe the only way to do the "mine for themselves" and compete successfully against traditional datacenters.

The distributed model competes by externalizing many of the costs by utilizing the scale of "mining as a hobby". Hobby miners don't count all their work hours as a cost. They don't always count all the hardware investments as cost. At best, they make some initial ROI calculation and make sure the income is greater than the cost of power. This has proven to work long before Bitcoin saw the light of the day. It started with SETI@HOME and then BOINC, which is now the largest super computer cluster on earth. Those projects didn't even have economic incentives.

The big technical problems to solve are related to security and hardware bandwidth. BOINC and other distributed computing have solved this but it is significantly less real-time than POW mining. Long lasting work packages and verification times in centralized systems is the state of the art. Typical low bandwidth mining rigs are not ideal.

A quote from the link...

"As far as the team is concerned, SNM’s current or future exchange rate is immaterial. They’re focused on building a marketplace for computing power, not an investment token."

This is exactly what I wanted to see. This tells us this project aims to produce wealth through actual use and is not just another pump & dump investment scheme.

Believe it or not but the entire development with ever more efficient hardware was anticipated already when Bitcoin was conceived.
POW is incentive based with greed as the driving mechanism and that is what makes it work. It's up to the blockchain projects to find technical solutions to problems with centralization. Social engineering as a solution can work short term but is not likely to work in the long run.

Luckily, it is comparatively easy and cheap to do software changes compared to develop new hardware. A blockchain project interested in keeping a certain type of hardware away should not have no problem doing so if there are competent developers.

Could you explain how the exchange mechanism works ?
Nicehash is efficient paying in BTC because there is no exchange involved. Other auto switchers fail because the long delay between profitability calculation/mining and exchange to BTC or other target currency.

What is the delay here ?

One useful feature that could mitigate the problem with rapid price fluctuations and exchange delay is to have the auto switching weighted towards difficulty priority instead of price priority.
If the profitability calculation makes a coin raise to the top because of a rapid swing in price, it may be worth less compared to if the difficulty drops for that coin. These two parameters are connected due to all bots and auto switching running out there so it's not easy to say what would work.
However, with some development effort, you could deal with this. You need to systematically collect statistics for a number of coins over time and save to a database. Then you can simulate different profitability strategies on that data and see what happens.

If you are really ambitious, you can use such data and feed it into an AI model. Then, the system could learn the market patterns for different coins and anticipate which coins will be most profitable to mine and when to send them to exchange. This may be the only way to beat Nicehash by a significant margin.

Having said all this, I thinks it's really good that you give Nicehash some competition.

Yes - 1

Meanwhile outside the sandbox...

There are actually people working on solving the problems with ASICs and centralization.
Everyone involved in mining would be wise to read up on these activities.

One line of work is a new type of POW algorithm called ProgPOW, which is designed to be much less profitable to implement in ASIC. Hopefully, we will see some projects adopting this.

Another track is the new movement with FPGAs. This type of hardware will fight with ASICs on technical merits and make it possible for the small miner to compete with big farms. The FPGAs can be reprogrammed when blockchains are forking or to mine different algorithms. They are also much more power efficient than GPUs, which will benefit small miners.

Also keep updated on which blockchain projects which are forking to keep ASICs out.
If you do not like ASICs, do not support blockchains that promote ASICs.
And - don't buy ASICs. If you want something more efficient than GPUs, look into FPGAs instead. There are multiple threads on this subject already.

A month or two.
I'm sure they will mail everyone who did the survey as soon as there are news

Sit back and wait...

The developers have a lot to do so the more they are distracted, the longer this will take.

You will continue to use 1x risers for the GPUs.
There will be special miner software that does the offloading to the accelerator.

Transactions like this is the perfect use case for blockchains and smart contracts.
Kind of funny we are not using the tech ourselves. Maybe some brainstorming is in order from the entire community on how to implement that in cases such as this.
I mean end-to-end, miner - middle man - Xilinx.

Would be an interesting exercise.

I think it's a mistake to mess with tokens and such complications.
Just implement the Nicehash business model and the project will take off like a rocket.

There are already too many distributed AI blockchain projects (DeepBrainChain, Neuromation, the Leonardo rendering thing etc.)  They make it so complicated just to get rich quick without taking any risks. Instead, focus on the actual use case and get something going as quickly as possible. Competition is coming so timing is crucial.

It's hard to do distributed AI due to the bandwidth requirements. I would make things simpler in order to get going more quickly. Set more stringent hardware requirements and let the "miners" deal with it. Look at what DeepBrainChain is doing. They promise a distributed solution but now they are selling high end, pre-built AI hardware and offer hosting. They are basically building a traditional datacenter and letting the "miners" pay for it. All this to buy time and make money quickly. They are selling thousands of these very high end machines (far above 1080 Ti) and the investors are lining up.

Instead, study the Nicehash example. They have made things very simple for sellers of hash power and thus attracted a huge amount of compute power. You could probably just log into their buyers marketplace and calculate the amount of GPU power they have available. Then make some estimation of how much of this is high end enough for AI. All that GPU power would switch to your solution if it would pay more and if it was equally simple to use.

From your question, it seems you need to read up a little more on how Nicehash works.
You will probably smile at your own question when you have realized what Nicehash is.

Question to the dev team:

Are there any plans for this miner when ZCOIN switches to MTP ?