Topic: Official FutureBit Apollo BTC Software/Image and Support thread - page 5. (Read 52508 times)

Difficulty:
Best share is the best solution your miner found.  Best share has to be above network difficulty to win the block.  Mining is broken up into levels of solutions that the miner can solve within a few seconds.  If you Just mine at max difficulty and it takes your miner minutes, then your miner could be mining an old block that has already been solved.  It's not the design to be notified when a block is solved but rather to question the network of the next block to solve.  Solutions are based on the transactions in each block making them unique.  The solution secures the transactions in the block as valid.

The error occurs when you submit a block solution that has already been solved or your node adjusts the diffiulty up or down and thus the miner solution doesn't match the difficulty.  Worst case is your mining on old blocks and have errors in your bitcoin network connection, high error rate.  Solutions found is to show how many solutions your miner correctly calculated.  Statistically speaking as your number increases over time you should have a better chance of solving the solution, Even though every single block the odds are the same.  IF you look at the solution of the Apollo II miner solving the block, the hash rate at that moment was Exahash in solving the solution.  Its why your miner constantly gives changing hash rate on every calculation.  The chance is you get lucky and your miner comes across the solution in a single calculation.  I"m at 1.7 billion best share.  I had a futurebit moonlander running for years and I had a best share of > 10 Trillion.  Close but not enough to solve the block.  It shows that even a 5 Watt futurebit moonlander can get close to solving the block.  That's the lottery I'll play over the lotto.  I think bitcoin odds are better.  All or nothing. 

My socket voltage is 240V @ 50Hz.
So 3500 / 240 = less than 15 Amps. Yes, I will run in turbo mode, so let's assume 4000 watts / 240 = less than 17 Amps. I got your point, I still won't run 10 from the same wire, only 3 to 5.

Ah, NO.  you can't run 10 units from a single household socket.  So, your running turbo.  That means at least 350-400 watts.  your little wattage meter is an indication of the watts being consumed by the Miner itself but does not include power consumption by the raspberry pi CPU.  350 Watts X 10 is 3500 watts.  3500 / 120 Volts = over 29 AMPS  you would be heating the wires in the wall.  Most household circuits are 15 or 20 amps. and you don't want to run max breaker power.  you want to run 75% of load for variance.  Running a 20 amp breaker at 20 amps will cause the breaker to eventually trip and fail.   You need to have an accurate digital tracking power meter to see how computers work.  IT's not a steady power but an average power.  The miner is constantly changing consumption multiple times per second and power fluctuates up and down a few watts per miner. 

You would want a dedicated 20 AMP circuit and even better 2-120 Volt circuits split across a 240 Volt phase at the panel.  That way each outlet balances out the 120V load if you have 5 apollos on each outlet. Don't forget a backup power supply and line conditioning regulation as the noise from 10 miners could cause problems.  Computers and switching power supplies are the noisiest additions to houshold ac circuits.

Thank you very much for your response!
You do not sound insensitive at all!

I do understand that Apollo units are literaly mini PCs. I have experience running servers 24/7 inside a dedicated AC cooled room. I will be running my Apollo inside an AC cooled room during a summer. However, during winter season it is more efficient to run it in a place that I do not heat up. My glassed balcony is covered, protected, clean and quite cool during winter. We use our balcony frequently everyday. It is ideal for Apollo unit as it won't have hard time cooling itself.

I will monitor Temp and Humidity and will ensure that the first is above 5 degrees C and the latter is below 90%. There is a very low risk to hit a Dew Point Temp.

It is great to know that I can run more than 2 Standard units via USB hub.
I think I can run at least 10 units from a single household socket, can't I?

I am excited to evelove further on in the home-based mining field!

I am pleased with Futurebit's response:

Hi temp ranges are from 5-40c, and humidity under 90% but making sure you are not operating near the dew point is what is essential, as this will cause condensation to form on the unit and cause damage.

As long as there is no head dust buildup on the unit, its intake or exhaust you should be fine, but will most likely require frequent cleaning to insure no dust settles on sensitive components.

You are not limited to two units and can run as many as your power/breaker allows with a USB hub.

Hello Apollo noderunners – working on an Apollo BTC issue where the OS is not detecting the NVME SSD. Occurrs on 2.0.5 and 2.0.6.

Tested a couple functioning NVME drives and neither is detected by the system.

Was able to see what seems to be a relevant problem listed during system startup...


Checking systemctl status

Has anyone encountered this before?

Forgive me for sounding insensitive: Are you for real? The Apollo units are mini computers and should be handled as such. If your intended location (glassed in balcony) provides a relatively clean environment other than being cold, use it. Obviously you are not going to use them "outside" where they may be subject to heavy moisture and/or heat. Your hounding of the Future Bit support staff for the "exact" tolerable criteria smacks of arrogance and/or ignorance. Do some research regarding pc hardware and you will easily find much general info.

Regarding the warranty: Future Bit clearly expects them to be used in the home (or home like environment). If maintaining the warranty is vital (for cost reasons, etc,) I get it. Most failures of new electronic goods will crop up very early in their lifespan. Run them a while and when comfortable they are "good", maybe worry less about warranty...

Again, searching with Google, DuckDuckGo, Ai, etc. will provide answers to your question regarding running multiple units from one node unit. I have read of several people plugging in a usb hub to the unit and hooking up several units that way. I run 6 Apollo units through one node unit wired through my home network. Research.

I can tell you from personal experience that the Apollo II full node can in fact run more than 2 standards unit.

Currently I have a total of 7 units running off the Apollo II full node. 4 x Apollo II standards and 3 x Apollo I BTC standards.

They replied:

I responded:
I will share their response.

My batting average for replies from FB support is 3-7 business days.  They aren't impressing me. 

And the switches are on back order (it took 2 emails and 7 days to get a response on this one), so I have no idea when my Apollo II will be running again.

I ordered another Apollo during the BF/CM sale.  They said 6-10 days.  Today is day 8, no notice of shipment yet.

just an update on this, the ddns on the route with no-ip and port forwarding did the trick without any problems.
if anyone is interested in solo mining with other apollos send me a dm

up.
I sent an email to Futurebit Support on the subject 2 days ago, but still didn't get an answer.

Correct

Ok, I am getting there.
Pools difficulty is just a handshake - makes sense.

Your "Best" should be > network difficulty for a valid block share - That means, that if out of a pure luck tomorrow early morning my Apollo finds a valid block, then I will see a number higher than 103T in my Best Share window of the Apollo Dashboard. Correct?

yeah that would work but with poor user experience.
You wouldn't get any stats until you submit a valid block share. For a solo pool like ckpool there is simply no other reason.

The pools difficulty is just a handshake in the stratum api.
For example:
you can send 1 share >100k with fluctuating stats.
you can send 10 shares >10k with smooth stats and balanced bandwith/ressource usage.
you can send 100 shares >1k with smooth stats and higher load
and without this limit you would (D)DoS the pool


As the others have said your "Best" should be > network difficulty for a valid block share.
Otherwise you can check that everything is working as expected (The Apollo generates a handful of accepted shares every minute).
Or just for fun to see how close your best share was but without any future correlation.
And you can see if your best share is statistically average or good/bad luck.

... each pool has its own smaller difficulty... - This one has just blown my mind. What does that even mean?

To my understanding the pool would check a single hash ONLY if it is submitted as a valid block. There is simply no other reason to validate a block.

I know that Apollo runs its own pool, but I don't understand how could that pool have its own smaller difficulty that is different from the bitcoin network difficulty? It makes no sense.

If by "the network difficulty" you mean the bitcoin network difficulty, then of course I understand this, as it is a core principle of the bitcoin network, but I cannot correlate it with the "Best", "Accepted" and "Rejected" Shares on the Apollo's Dashboard.

I updated to 2.0.6 today and the apollo-api service no longer starts. Anyone got any tricks to debug?

All I get in syslog is

systemd[1]: Started Apollo API & UI app.
systemd[1]: apollo-api.service: Main process exited, code=killed, status=7/BUS

And running the systemctl command myself

sudo /usr/local/nvm/versions/node/v21.6.2/bin/node /opt/apolloapi/src/init.js
Bus error


Seems its still mining, but the UI is broken.

yeah so I have 8333 and 3333 already port forwarded to my apollo node
weird that the 3333 port showed it was close yesterday but now it shows open, I guess all I have to is test now - will update when I get my results

thanks

That is correct for the bitcoin network difficulty.
But each pool has its own smaller difficulty for each miner because it is neither possible nor necessary for the pool to check every single hash.
For solo mining the Apollo runs a "mini pool" in the background.

The network difficulty increases as more/faster miners find valid hashes.

Yes it should be if your router/firewall is set to allow it. But I think that is why we have to use 8333 port forward is it allows the other nodes to connect to our node? I am sure there are others hee that understand it all better than me