10 March 2018
MF
01:27
Mr Flintstone
1.8 th/s over last 20 blocks
P|
01:34
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
Awesome update! Thanks man!
01:34
😍
Z
01:37
Zack
In reply to this message
This is unrelated to governance variables.

Node operators can make money by facilitating lightning payments and smart contracts.
01:40
In reply to this message
Did you see the white paper? We do turing complete smart contracts in the state channels.
01:42
In reply to this message
Probably your database is corrupted and you need to resend the blocks.
J
01:43
Jim
Yes, a tutorial / demo on setting these up can help. Will watch the YouTube video.
Z
01:45
Zack
In reply to this message
The miner can request new work as quick as they want. I didn't write it.
The pool I wrote checks every 3 seconds for new work.
The full node generates new work if a block is found, or if more than 40 seconds or so have passed.
MH
01:46
Mandel Hoff
Does the node generate new work if mempool changes or anything with timestamps?
Z
01:47
Zack
In reply to this message
It has a 40 second timer. In potential_block.erl
01:47
Or maybe 60 seconds
I
01:48
Iridescence
@Mandelhoff Are you seeing the bhash change much more often than expected?
MH
01:49
Mandel Hoff
No, it's pretty much what Zack said now. I think it just used to change more frequent when block times were super-sonic.
01:50
Based on those past observations, I was entertaining some incorrect assumptions and consequences of bhash changing based on certain events.
OK
01:56
O K
In reply to this message
Where is memory controller load statistic found? I have noticed no significant stutter on any release
MH
02:03
Mandel Hoff
That TechPower GPU-Z tool for Windows shows good GPU monitoring metrics. I watched it there. If you don't have OS stutter, you will almost certainly want to leave suffixMax at 65536
OK
02:09
O K
No other reason to move from 6 to 7 right?
MH
02:09
Mandel Hoff
Go to 7. It's way faster.
OK
02:10
O K
Oh okay πŸ˜„
P|
02:11
Paul N. | BAND (πŸ…‘) - I'll never DM you
@Mandelhoff your pool down?
MH
02:11
Mandel Hoff
Pool down. Fixing. Node died with "died" errors.
M
02:14
Minieep21
wtfffff the speed! Thank you Mandel!
MH
02:14
Mandel Hoff
πŸ‘
P|
02:14
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
😳
MH
02:14
Mandel Hoff
I'm looking for 1080 speeds on the new miner if anyone has speeds and optimal blocksize/numblocks for a 1080 please share.
M
02:21
Minieep21
2.2 GH/s on 1080 ti with 96 168
MH
02:21
Mandel Hoff
Thank you.
OK
02:21
O K
I have 1080 but
02:22
Only one miner will work, other 0 h/s
M
02:22
Minieep21
Default settings got to 1.9 GH/s
OK
02:23
O K
It was some argument I set
02:23
I wish the random string was before the other arguments so I could leave them default
MH
02:24
Mandel Hoff
I updated the github readme for the 1080ti best settings. Thank you.
OK
02:25
O K
Do we still need the random string for multiple cards
M
02:25
Minieep21
Tested SuffixMax from default to 1,000.
Stuttering continued to be a problem until about 5,000 for me. Currently running at 1,000 SuffixMax with stable 2.16 GH/s
MH
02:25
Mandel Hoff
Yes, the argument handling could use a refactor.
02:25
You don't have to use random string - but, if you get "duplicate nonce" errors then do use random strings.
OK
02:25
O K
I'm getting 1.3 on dfault settings on my 1080s
02:25
Between 1.25 and 1.35
02:26
But if I change any settings only one card will work
MH
02:26
Mandel Hoff
The pool protects against duplicate nonce attacks, so it's pretty obvious by that error if you need the random string seed or not.
OK
02:26
O K
Only since 65536 argument address
02:26
Added*
J
02:27
Jim
96 168 1000 gives 1.68 Ghs on 1080 (modest 100 MHz core overclock)
OK
02:27
O K
My 1070 started at 1G but slows now, my cards always slow from fresh start
MH
02:28
Mandel Hoff
In reply to this message
πŸ‘
M
02:30
Marc
where can I find the log of the amoveo-mining-pool? it does not talk very much.
02:30
or how can I be sure it is running fine?
J
02:33
Jim
96 256 1000 gives me 1.3 on a 980 ti
02:33
Really massive improvement for both here
02:33
Will keep tweaking
02:35
Deleted Account
total network hashrate?
MH
02:35
Mandel Hoff
The suffixmax 65536 is for sure the best speed, but it can lag the OS pretty bad at that high.
02:36
In reply to this message
1.2th/s
OK
02:38
O K
Is there any reason why the entropy shouldn't just be added behind the scenes? Does it have a negative impact on people mining on one card?
MH
02:38
Mandel Hoff
I keep reading Augur's "weekly reports" thinking they should build their market features on top of Amoveo instead of Ethereum. πŸ‘ Great job Zack.
02:40
In reply to this message
Someone asked me to put the seed string on the command line for scripting multiple gpu's, so I did that. It used to prompt for user input, and I didn't like that.
OK
02:40
O K
No I don't mean that
02:40
I mean, without any user input or scripting at all
02:42
In reply to this message
Like this?
MH
02:42
Mandel Hoff
I would use tool launch time or something like that. I had pseudo-launch time seed in the cpu miner.
P|
02:43
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
That's a line in my batch script for starting the miner
OK
02:43
O K
I'm sorry I'm not communicating clearly
MH
02:45
Mandel Hoff
I'm using the c++ random bytes generator to create nonces, but it's terrible at seeding itself, so user input or launch time or some external seed string works well.
OK
02:45
O K
That makes sense, I understand now
MH
02:45
Mandel Hoff
πŸ‘
Z
02:48
Zack
In reply to this message
If you can connect miners to it, then it is working.
M
02:48
Marc
Ah. Ok. Thanks.
Z
02:48
Zack
In reply to this message
http://159.65.120.84:8080/explorer.html
click the "height" button.
I
02:49
Iridescence
somebody should do amoveostats.com
02:49
Z
02:50
Zack
In reply to this message
cool webpage. I hadn't seen this before.
Ghadras joined group by link from Group
G
02:59
Gonzalo
In reply to this message
I confirm this stats, well done @Mandelhoff !!!
MH
03:00
Mandel Hoff
πŸ‘ Thank you.
G
03:00
Gonzalo
SO is very laggy and after a while my 2 1080ti are dropping to 211 and 197, but anyway, nice improvement
03:01
maybe is because Im connected with RDP to mi miner
03:01
This second testnet has been quite productive, I guess we are prepared to go for the MAINNET πŸ˜‚
MH
03:02
Mandel Hoff
I keep thinking that same joke, but it makes me a little sad inside too.
G
03:02
Gonzalo
haha anyway, I dont believe this would change things too much
03:03
not now
03:11
testing: GTX1060 6GB BlockSize: 64, getting 875mhs πŸ‘
BB
03:13
Brian Brian
In reply to this message
Wow I get 250 mh/s on Ubuntu but I’m also using the card for graphics. Will change my setup around. Do you guys think I should be able to reach these speeds with Ubuntu?
G
03:18
Gonzalo
at current hashrate, how many veo will get 1ghs?
OK
03:19
O K
In reply to this message
The 1000 slows me way down, but considerable improvement over stock for 96 168 on 1080 and 1070
03:19
1.5-1.6 for me
03:19
and 1.2 for 1070
Z
03:19
Zack
In reply to this message
A block is 340 TH. so if you have 1 gh/s, then you need 340 000 seconds to find a block.
03:20
about 4 days.
P|
03:20
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
which is ~4 days
03:31
Deleted Account
AmoveoMinerGpuCuda's linux port updated to 1.0.0.6
https://github.com/tumblecatweed/AmoveoMinerGpuCudaLinux
If you update original repo's code to 1.0.0.7 I'll try @Mandelhoff
MH
03:33
Mandel Hoff
Ok, one minute for code push.
P|
03:33
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
Can you add the dependencies?
03:34
I'm giving it a try
03:34
thanks :D
03:34
Deleted Account
ok
03:38
Added dependencies
P|
03:39
Paul N. | BAND (πŸ…‘) - I'll never DM you
Awesome! Thanks man
MH
03:40
Mandel Hoff
All of the 1.0.0.7 code is on github now.
G
03:42
Gonzalo
In reply to this message
oh I forgot info was on the new explorer, thanks Zack πŸ‘
sd
03:45
steel dan
Any news about max supply and emission time line? We can't price amoveo if some upper limit is not set...
03:46
Deleted Account
MH
03:47
Mandel Hoff
Great job catweed !
03:47
Deleted Account
This one is really fast!
03:47
amazing
Z
04:03
Zack
In reply to this message
The block reward is controlled by the governance mechanism. I cannot predict what the community will decide to do.
04:04
Trying to predict an ideal emission timeline years ahead is a recipe for failure.
We need to adapt to survive.
sd
04:04
steel dan
Yes but until that moment, all pricing attempts are pointless
P|
04:05
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
Not exacly. Veo can be priced on mining cost.
sd
04:06
steel dan
Given that cryptos are divisible there is no issue to dictate 10**9 units for instance and let the market adjust
04:06
Any arbitrary number is OK if it is known a priori
04:07
Then the market can price amoveo market cap at 10 or 50 billion dollars for example and price discovery is automatic
04:07
Scarcity is precious to incentivize market participants to assign value and trust a cryptocurrency
04:08
If value is uncertain people would not be able to use amoveo as a currency to place bets
04:09
So IMHO a supply cap should be dictated as soon as possible
04:10
And maybe allow some small diminishing inflationary emission just like monero to incentivize miners after the main emission is complete
04:11
Just copy monero emission scheme would suffice
Z
04:11
Zack
I can barely plan my personal budget for next month.
You expect me to plan the budget for an entire cryptocurrency, years ahead of time?
04:11
Sounds like a bad strategy to me.
sd
04:11
steel dan
Don't worry about it, market and hashing power will adjust
04:11
The invisible hand
04:11
Veo is divisible
04:11
So no problem
04:12
If market cap spikes people will use nano veo or micro veo currency units
Z
04:12
Zack
The market can only optimize one variable.
We are trying to optimize for 2 variables here.
sd
04:13
steel dan
For instance Bitcoin scarcity has made it so sexy and people got heavily engaged with it
04:14
Even if most of it is locked up in wallets the remaining free floating Bitcoins are enough to drive the crypto economy
04:14
BTW only scammy coins use large max supplies
04:15
... Tron, pac etc
04:15
Verge...
s
04:15
scott_l | DeFi Pulse
In reply to this message
bitcoin is alot of things, but sexy is not one of them
sd
04:15
steel dan
In reply to this message
I mean the passion of people to hodl it
04:16
What can kill cryptos in general is the oversupply of forks, alts, clones, tokens etc...
04:17
It triggers inflation and makes the entire crypto market a joke
MF
04:17
Mr Flintstone
@stjor what makes you think the bitcoin block reward isn’t a governance variable?
sd
04:17
steel dan
Which is sad because ignorance of the people makes all cryptos highly correlated
MF
04:17
Mr Flintstone
the distinction here is on chain vs off chain
sd
04:18
steel dan
In reply to this message
It takes a hard fork to change it ok, but noone dares to open that issue
Z
04:18
Zack
What stelios suggests would make it so that the market cap and block reward are always in proportion.
That is not ideal.
In that situation, Amoveo would be unable to grow quickly when small, and it would be unable to reduce the growth rate once it is big.

I think that the block reward and market cap should not be in fixed proportion. They should be free to move relative to each other.
MF
04:19
Mr Flintstone
In reply to this message
so it is apparently that people like scarcity right?
04:19
Apparent*
sd
04:20
steel dan
In reply to this message
It gives you confidence to hodl
MF
04:20
Mr Flintstone
people in general want their asset to be scarce. I think amoveo will be no different
04:20
regardless if the block reward is chosen off chain or on chain
04:20
people are free to disagree
sd
04:20
steel dan
In reply to this message
You can fork Bitcoin core and go on your own of course with a new emission scheme if u dare but then it's just a scammy Altcoin u have spawned
MH
04:21
Mandel Hoff
I don't view Amoveo as a currency as much as it is a platform for markets. I'm not sure scarcity applies in exactly the same manner as something that is "just a coin".
s
04:21
scott_l | DeFi Pulse
In reply to this message
this 100%.... amoveo is a platform
04:21
10 years in, bitcoin still moving 5-10% on big days.
sd
04:21
steel dan
In reply to this message
Given that amoveo reached 340 Tera hash in a few days it could be the Ethereum killer
s
04:22
scott_l | DeFi Pulse
VEO cant have moves that big, because it is a unti of account for its market
04:22
markets
sd
04:22
steel dan
Amoveo is definitely more than a coin but it should also entail the properties of a coin
MH
04:22
Mandel Hoff
Amoveo doesn't do ICO's, so Ethereum can keep its day job. πŸ˜†
MF
04:23
Mr Flintstone
In reply to this message
You can hedge your veo exposure with the oracle tho
Z
04:23
Zack
In reply to this message
sure it can. Amoveo can make financial derivatives to protect you from movements in the price of Veo.
MF
04:23
Mr Flintstone
I’m not sure volatility matters
04:23
exactly
s
04:23
scott_l | DeFi Pulse
In reply to this message
are u envisioning people betting on amoveo markets with collared off VEO?
sd
04:23
steel dan
In reply to this message
If amoveo price is robust then it could also be used for crowdfunding, anything goes
Z
04:24
Zack
Amoveo is a currency.
Like all other currencies, it is a bad idea to make the rate of issuance be a fixed ratio to the market cap. Because a larger market cap does not necessarily mean a higher demand for new tokens.
04:24
In reply to this message
You can use Amoveo for dominant assurance contracts, which is a form of debt-free crowdfunding.
You can't use Amoveo smart contracts for any debt.
MF
04:25
Mr Flintstone
In reply to this message
what exactly do you mean by debt?
sd
04:26
steel dan
In reply to this message
Yes but to emulate financial derivatives you need solid value for the payoff in a unit resembling a currency, that's how people can hedge their real life risks, so any actions towards making the amoveo price solid as much as possible makes the amoveo platform more efficient
Z
04:26
Zack
In Amoveo, everyone is always holding their own money.
MH
04:26
Mandel Hoff
In reply to this message
You're right ICO's are sort of a crowdfunding. I like Eth very much - was just a joke.
sd
04:27
steel dan
I'm not saying we need to kill volatility no matter what...
Bitcoin or Ethereum volatility is acceptable... But volatility of shitcoins from cryptopia would not be acceptable :)
MF
04:28
Mr Flintstone
what decreases liquidity is when people provide liquidity to the market
Z
04:28
Zack
One major reason that cryptocurrencies are so volatile is because of the lack of financial derivatives on cryptocurrencies.
Amoveo will allow for the these derivatives in secure trust-free market places. This will make all cryptocurrencies more stable.
MF
04:28
Mr Flintstone
decreases volatility*
sd
04:29
steel dan
To make amoveo useful in as many use cases as possible is essential
MH
04:29
Mandel Hoff
In reply to this message
Can I retweet this? It's excellent. πŸ‘
Z
04:29
Zack
sure
MF
04:30
Mr Flintstone
In reply to this message
it’s important that the notional exposure of the financial derivative is collateralized in the spot market in order to decrease volatility though
04:31
derivatives by themselves are not enough
Z
04:31
Zack
You can choose whatever margins you want on your contracts. it is turing complete.
04:31
Maybe I don't understand
MF
04:32
Mr Flintstone
zach, if we bet on the price of veo in the future, just between you and me
04:32
this is a financial derivative
04:32
but it doesn’t help with liquidity unless one of us goes out into the spot market and hedges our exposure
04:33
large banks provide this function in traditional capital markets
04:34
they provide liquidity for derivatives in exchange for a fee ( i.e take the other side of the derivative bet), then collateralized that bet in the spot market in order to make sure they make the fee spread without any actual exposure to the underlying
Z
04:34
Zack
The word "liquidity" doesn't make sense to me in this context.
That word mean: A more liquid market is one where the price moves less when you make bigger trades.

Are you talking about the liquidity of the spot market? It is not Amoveo's goal to make other markets better.
MF
04:35
Mr Flintstone
I’m talking about the liquidity of the amoveo spot market
Z
04:35
Zack
There are lots of uses for derivatives besides what you describe.
Like if I need to own a bunch of Ether temporarily, I could use Amoveo to hedge against the risk of Ether losing value.
04:36
Does "spot market" mean the cryptocurrency exchanges, like trading Veo for Bitcoin?
MF
04:36
Mr Flintstone
yes
Z
04:37
Zack
The smart contracts on Amoveo have margins programmed into them. The contract is fully collateralized to the margins.
04:37
Otherwise it would be debt.
04:37
Amoveo cannot do debt.
MF
04:38
Mr Flintstone
maybe I am not clear in what I am saying
04:38
it is this: financial derivatives will help with decreasing volatility, but not all financial derivatives
04:39
I probably could have just said that instead of writing a time
04:39
Tome*
Z
04:39
Zack
yes. for example a derivative that is unused doesn't accomplish anything.
Z
04:58
Zack
I see 5 different miners making recent blocks
04:59
I guess 2 are solo miners, they didn't pay rewards
s
04:59
scott_l | DeFi Pulse
In reply to this message
i think financial derivatives will help in some instance, but i was worrying about the case where i bet say 20 VEO that trump will win in 2020, i cant hedge the VEO from that bet bc it is out of my control until i close my bet or it settles
05:01
unless im misunderstanding the system
Z
05:02
Zack
In reply to this message
It is probably too expensive to leave money in channels that long.
Especially at first, Amoveo it best for bets that will resolve within about 2 weeks.
05:02
Or less
05:03
But you can hedge Veo that is being bet.
You can price derivatives in other derivatives. So a bet on the presidential election could be priced in stable coins.
07:19
Deleted Account
total net hashrate?
MF
07:23
Mr Flintstone
You can see on light wallet now
MF
07:31
Mr Flintstone
@zack are you still running into that 80 second issue with the node?
BB
07:32
Brian Brian
Wow mandel I have a 10x improvement with the latest code. TYVM!
Z
07:32
Zack
No. It was fixed
07:32
In reply to this message
Awesome! Good job Mandel
MH
07:34
Mandel Hoff
Thanks. I'm glad it's working for you πŸ‘
Z
07:56
Zack
@Simon3456
I created a transaction using my light wallet.
I see that it appeared in my mempool on the server.
Someone else mined a block, and they did not include the tx.
The tx is still in my mempool.
I manually shared the tx with Sy's server. His explorer shows that my balance changed.
I check Gonzalo's server. His explorer shows that my balance has not yet changed.

It seems that the manual tools for sharing txs are working correctly, but they are not happening automatically.
S
07:59
Sy
Thanks for tearing
07:59
Testing
Z
08:02
Zack
Mandel mined a block that Sy and I are refusing to download.
S
08:05
Sy
I am? πŸ˜…
Z
08:05
Zack
I confirmed a third node that is refusing to download Mandel's block.
S
08:05
Sy
huge payout from BKyL
OK
08:05
O K
What is the third node
MH
08:05
Mandel Hoff
Accidental fork?
Z
08:05
Zack
Mandel has 2 bad blocks
08:06
yes, Mandel has like 23 transactions in the block that we aren't accepting.
MH
08:07
Mandel Hoff
What to do? Clean and sync over?
S
08:07
Sy
yeah my node is going crazy
08:07
potentiel block died etc
Z
08:08
Zack
In reply to this message
yes, I think your database must be corrupted.
Did anything weird happen while you were running this time?
08:08
sounds like Sy needs to clean the blocks and resync too.
S
08:09
Sy
make prod-clean right?
Z
08:09
Zack
yes
MH
08:09
Mandel Hoff
Do I need to clean if I have bad blocks?
Z
08:09
Zack
my node keeps searching for the blocks to go with the headers I got from Mandel. but it can't find any valid blocks
S
08:09
Sy
resyncing
08:10
did mandel kill the chian? xD
I
08:10
Iridescence
Why are you refusing to download the block?
08:10
Node bug?
Z
08:10
Zack
In reply to this message
yes. I think that the only way you could get bad blocks is if the database is corrupted.
You might want to keep a record of the txs we are undoing.
08:10
light nodes wont work until we mine 3 blocks to recover from the fork.
G
08:10
Gonzalo
both my 2 nodes are at 6593 headers & blocks: 51.15.75.100 and 51.15.69.135
Z
08:11
Zack
oh, then Maybe Mandel is healthy
MH
08:11
Mandel Hoff
Good catch Gonzalo.
08:12
There are 3 different nodes that wont download Mandel's blocks.
S
08:12
Sy
syncing...api height tells me 6593
OK
08:12
O K
Mine is 6592
S
08:12
Sy
4400....
MH
08:13
Mandel Hoff
Ok, I'll stay "on" for a couple minutes to see if you can sync to my height. The 23 people getting payouts would love that block to go through. πŸ˜†
G
08:13
Gonzalo
S
08:13
Sy
magic :D
G
08:14
Gonzalo
In reply to this message
πŸ‘πŸ‘πŸ‘ yes!
08:14
go Mandel, go!!
08:14
πŸ˜‚
S
08:14
Sy
Mandel your new Title from now on is "Node Killer" xD
MH
08:15
Mandel Hoff
I want stable nodes too. Feels like I can't go to sleep anymore. πŸ™ˆ
S
08:15
Sy
6593...any node further ahead?
G
08:15
Gonzalo
In reply to this message
wanna mine on my pool for a while?? 😁
08:16
In reply to this message
oh :(
S
08:16
Sy
In reply to this message
Sticker
Not included, change data exporting settings to download.
πŸ˜‚, 36.1 KB
G
08:16
Gonzalo
hjaha
MH
08:16
Mandel Hoff
Did you absorb my 6593 successfully?
S
08:16
Sy
okay i just got 6594
08:16
guess its working again
Z
08:16
Zack
Sy downloaded Mandel's block
S
08:16
Sy
and 5
G
08:16
Gonzalo
me too, and 6595 as well
MH
08:17
Mandel Hoff
Awesome! Big payday!
S
08:17
Sy
yep...dont do that again πŸ˜‚
MH
08:17
Mandel Hoff
I had a huge back log of txs from ones that got lost. It probably won't be like that again - maybe.
S
08:19
Sy
i wonder if it would have fixed itself if i just waited a bit longer....
Z
08:19
Zack
I am resyncing to get onto the same fork as you guys.
I wonder what went wrong.
Maybe it had something to do with me sending a tx to Sy's server?
08:20
In reply to this message
http://203.80.236.230:8080/explorer.html
This node is still stuck. So I think it does not fix itself.
OK
08:20
O K
I'm still at 6592
Z
08:20
Zack
In reply to this message
do you mind testing something?
MH
08:20
Mandel Hoff
I crashed about 2 hours ago where my "restart" was ~20 blocks behind where it crashed at - probably not related, but worth noting.
OK
08:21
O K
no please let me help
Z
08:21
Zack
In reply to this message
tx_pool:get().
08:21
Which txs are in your mempool?
OK
08:21
O K
I just logged in, I'm seeing lots of block absorber died!
08:21
sec
08:22
How can I output this tx_pool:get(). somewhere readable, it scrolls too quickly
S
08:22
Sy
im off to bed, read you all tomorrow
Z
08:23
Zack
In reply to this message
When did you last update? there should be a lot less spamming messages now
OK
08:23
O K
sec, I'm going to scroll up and close my pool and use CTRL+insert to copy
08:23
yesterday evening I think
Z
08:23
Zack
Thanks for trying anyway. don't worry if it is too much trouble
OK
08:23
O K
Is there an uptime function
Z
08:23
Zack
that is pretty recent.
08:25
once more. I created the tx from the light node, and I see it on the server.
08:26
I pushed it to both Mandel and Sy.
08:27
using the explorer i see it appeared in both their mempools.
OK
08:30
O K
08:32
I have NO idea why it has a .crash extension, telegram did that
08:32
Let me know if I can pull anything else from it before I take it down and get synced
Z
08:33
Zack
looks like you have a reward tx that my node made.
Thank you for the info
08:33
I don't think we need anything else from your node
OK
08:34
O K
Glad I could help
Z
08:35
Zack
this time my tx got included in a block mined by Sy.
MH
08:35
Mandel Hoff
That sounds like great results
Z
08:44
Zack
https://www.reddit.com/r/AmoveoTrader/comments/83bnlt/auction_of_002_btc/
I am doing an auction in the other direction this time.
I think this will give us a lower limit on the price of Veo. The previous gave us an upper limit.
09:29
Deleted Account
@Mandelhoff is there a minimum payout on your pool?
G
09:30
Gonzalo
1 veo
09:30
Deleted Account
right, thanks Gonzalo couldn't see that listed anywhere πŸ™‚
G
09:31
Gonzalo
πŸ‘ I think (I hope!) it will be less in the future
RT
09:31
Roger That
i think that payout will be deemed very high as this hashrate goes up exponentially
N
09:31
NM$L
09:31
1080
09:31
nice hashrate
09:31
on Linux
G
09:32
Gonzalo
πŸ‘
N
09:34
NM$L
1.3G
MH
09:35
Mandel Hoff
πŸ‘ Awesome. I'm glad it got ported to Linux too.
N
09:35
NM$L
In reply to this message
get found share
09:36
MH
09:36
Mandel Hoff
Yes the 1.3 ghs is a valid rate with the latest version. It's very fast.
N
09:36
NM$L
πŸ˜‹
G
09:36
Gonzalo
so 4 days=1 VEO
09:37
*if hashrate remains the same, of course
09:38
Deleted Account
so the explorer at http://159.65.120.84:8080/explorer.html says ~1.7TH nethash and Mandel's pool says ~64TH... wondering why the massive discrepancy? πŸ˜„
09:38
and that's under "pool hash rate" not network hashrate?
09:38
now down to 3.85TH, I guess it is busted
09:39
but is 1.7TH a good estimate?
09:39
maybe an average vs instantaneous?
MF
09:42
Mr Flintstone
In reply to this message
1.7 is average of last 20 blocks
09:43
Deleted Account
right, thanks
Z
09:44
Zack
The estimate program I made for the explorer is probably correct.
Arithmetic can be challenging.
MF
09:51
Mr Flintstone
@zach are there any live markets right now
Z
09:52
Zack
In reply to this message
no.
MF
09:52
Mr Flintstone
zack* sorry
09:54
maybe it makes sense for us as a community to start familiarizing ourselves with how to interact with these things at some point soon, esp because they are used for governance
09:54
I know there are people here that know exactly how to use them, but I would guess a lot of people myself included do not
Z
09:54
Zack
In reply to this message
The on chain oracle is different from the off-chain markets.
MF
09:55
Mr Flintstone
oh interesting
Z
09:55
Zack
The markets are a program written in the turing complete smart contract language. They are inside the lightning network
09:56
The on-chain oracle uses transactions that need to be included in a block.
09:56
You can do all the market stuff from the light node.
The Oracle stuff has to be done from the command line for now.
09:57
I expect that almost everyone who uses amoveo will participate in markets.

But very few will participate in oracles.
MF
09:57
Mr Flintstone
why do you think?
Z
09:59
Zack
So let's say we are gambling on a football game.
The markets would let us make bets before and during the game.
The Oracle only lets us make bets after the game is over and the result is known.
Most oracle bets will go unmatched, because no one wants to bet against reality.
10:00
If you are betting against someone in the Oracle, then that means one of you is probably maliciously attacking amoveo.
MF
10:00
Mr Flintstone
interesting
MF
10:01
Mr Flintstone
lol
10:01
are all oracle bets recorded on chain?
Z
10:02
Zack
In reply to this message
Yes. All oracle interactions, including bets, are on-chain transactions.
MF
10:02
Mr Flintstone
that’s awesome
10:03
so the markets pay out based on what the oracle says?
Z
10:03
Zack
In reply to this message
Docs/design in the amoveo github has some stuff about oracles
Gnosis "ultimate oracle" is a similar design.
Paul sztorc wrote a lot of great info about oracles, but our design is very different from his. He is doing something similar to augur.
10:04
In reply to this message
The markets smart contract I wrote is connected to 1 oracle each. The result of the market is determined by the Oracle.
MF
10:04
Mr Flintstone
cool
Z
10:04
Zack
In the future we will have market smart contracts that are aware of more than one oracle.
MF
10:05
Mr Flintstone
can you have one oracle that reports the price of the S&P 500 at the end of each day
10:05
or would you need multiple oracles, one for each day
Z
10:06
Zack
In reply to this message
No. The Oracle only answers yes/no questions.
You could have each oracle tell you one bit of the price encoded in binary. Or you can ask the Oracle "will the market be above price X?"
MF
10:06
Mr Flintstone
actually are the oracles only answering yes/no?
10:07
o :(
10:07
having one oracle for each bit in binary is a genius way to get around this issue wow
10:07
I never thought about that
10:08
so you could have an oracle system that only reports the price of the sp500 once you interpret the bits that are reported
10:09
Deleted Account
guess I'd better stick with markets for taking a punt on the outcome of 1000 monkeys on typewriters then πŸ˜„
Z
10:09
Zack
The smart contract would need to be aware of all the different oracles. The smart contract would interpret them as binary, and transfer an amount of money based on that binary number.
MF
10:09
Mr Flintstone
yep, but this wouldn’t be hard to do
10:09
I wouldn’t think so at least
Z
10:11
Zack
In my experience the actual writing of the smart contracts has been easy.
But passing messages back and forth between the participants in each smart contract is challenging.
It is hard to remember to verify all the right things and it is hard to know the correct order to move the data
10:12
I already wrote a market and a lightning payment contract.

I feel that these two tools are fundamental, and all the smart contracts we make will be variations on them.

Hopefully we can reuse the protocol for passing messages as much as possible.
10:14
Deleted Account
very cool tech & vision, Zack ... how are you funding the effort?
MF
10:14
Mr Flintstone
very cool
Z
10:15
Zack
My expenses are very low. I haven't hired anyone.
The community has been a lot of help.
10:16
After that auction, I think I have gained more money than I spent while working on Amoveo.
MH
10:16
Mandel Hoff
πŸ‘ Zack is an army of juggernauts by himself.
MF
10:16
Mr Flintstone
we also have the developer reward
Z
10:17
Zack
Yes. If you include Veo tokens, then I definitely gained more than I lost to this project.
10:18
So it is a profitable project. I don't need funding.
P|
10:22
Paul N. | BAND (πŸ…‘) - I'll never DM you
Is there any reason miners shouldn't try to increase veo payout proportionately to the difficulty change of the last adjustment?
MF
10:23
Mr Flintstone
In reply to this message
some miners might want veo to be scarce
Z
10:23
Zack
In reply to this message
Mining tech is improving rapidly. The cost per hash is decreasing.
10:24
We are something like 200 000x more expensive per hash than bitcoin at his point.
MH
10:24
Mandel Hoff
Pretty certain we found a gpu peak. Next leap will likely be community scaling or fpga/asic
N
10:25
NM$L
what's the max supply
MF
10:25
Mr Flintstone
In reply to this message
this is not determined
10:26
because the block reward can change
N
10:26
NM$L
I have 1 veo
MF
10:27
Mr Flintstone
nice! I remember you had 0.6 yesterday
10:27
so that is a good day of mining
N
10:27
NM$L
I have totally 1.5G hashrate now. Not bad.
Z
10:28
Zack
I think that the hashrate will increase a bunch once we settled our first few markets and smooth out the details to make it easy to participate in markets.
This will give miners the confidence to increase their investment.
10:29
We need to demonstrate the killer app of amoveo
N
10:29
NM$L
when sell is most profitable
MF
10:29
Mr Flintstone
In reply to this message
I agree
N
10:31
NM$L
Sell my 1 veo at 1 ETH/Veo
JS
10:32
Jon Snow
Sell my 1 veo at 0.9 Eth
Deleted joined group by link from Group
Z
10:33
Zack
There are a lot of problems with the software.
Half the network died today.
Be careful investing in Amoveo. You could lose everything you invest.
P|
10:34
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
πŸ˜‚
N
10:34
NM$L
10:34
One day in the future
P|
10:35
Paul N. | BAND (πŸ…‘) - I'll never DM you
Zack You can be sure I'm not being careful at all 😭
10:37
I guess I made a big mistake selling for only $54 each
P|
10:39
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
You didn't, people who didn't participate in that auction did.
Z
10:41
Zack
If Amoveo is really over $93, then that means our market cap is over $850k
N
10:41
NM$L
get list in CMC
Z
10:42
Zack
We probably need a little more trading volume first
P|
10:42
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
I don't think that's a good idea
MF
10:42
Mr Flintstone
In reply to this message
yeah I agree
10:42
we really need the network to be stable
P|
10:42
Paul N. | BAND (πŸ…‘) - I'll never DM you
we need some more tools/tech around this project to be stable
N
10:43
NM$L
now there are only 6648 veo?
Z
10:43
Zack
9426 veo in total
MF
10:43
Mr Flintstone
what about not including the developer reward?
N
10:43
NM$L
1 veo per block
10:43
why 9426
Z
10:44
Zack
7855 without developer reward.
10:44
the block reward was 1.5 for a while.
10:44
then we changed it back to 1 again.
MF
10:44
Mr Flintstone
lol the developer reward is about to be a serious stack
P|
10:44
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
It already has πŸ‘
MF
10:44
Mr Flintstone
but the community controls it?
10:45
at least who it is paid to, and in what amount, and how frequently?
P|
10:45
Paul N. | BAND (πŸ…‘) - I'll never DM you
Can we make a bet to steal that?
10:45
😏
RT
10:45
Roger That
Don’t list. Let this evolve. Let’s not jump through hooves.
Z
10:45
Zack
the governance mechanism can change how much I am getting paid. It cannot change who the reward goes to.
The plan is to eventually switch to insured-crowdfunding to pay for development.
10:46
I cant access the funds until about 12 months in.
RT
10:46
Roger That
Discord has 1 veo = 1 bitcoin from sellers. Not sure if they are serious.
MF
10:47
Mr Flintstone
lmao I’m pretty sure they are not
P|
10:47
Paul N. | BAND (πŸ…‘) - I'll never DM you
I think they, if you're willing to pay
10:47
they are*
RT
10:47
Roger That
Lol
10:47
Yeah. Would be cool though πŸ˜‚
P|
10:48
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
Are you willing to man?
MF
10:48
Mr Flintstone
In reply to this message
can we make a smart contract that pays people if they succeed in making the network more stable? I feel like we should have more people than just you looking at this and trying it fix it
P|
10:48
Paul N. | BAND (πŸ…‘) - I'll never DM you
I am willing to sell at 0.9 BTC/veo
10:48
cheap coin really
RT
10:49
Roger That
Sell yes. Buy no as I am not rich enough. Maybe Kassel can. Corner the market totally.
MF
10:49
Mr Flintstone
although I guess it is already in our self interest to make the network stable
Z
10:49
Zack
In reply to this message
Yes. they should write the code first, then explain how it works, then we make the contract, then their release the code as open source to get paid.
10:50
Lets make some other markets first though.
10:50
I am thinking the first market should be on the future difficulty of mining a block.
MF
10:50
Mr Flintstone
In reply to this message
I think this makes sense
10:50
In reply to this message
Difficulty of block 8001?
Z
10:51
Zack
I am nervous to put too many Veo into a market server. So the first market wont have much volume.
10:51
one we do more testing, then we can slowly make bigger markets.
MF
10:51
Mr Flintstone
In reply to this message
I assume you can easily limit the amount that allowed to be bet
10:51
In reply to this message
But someone could just fork it
Z
10:52
Zack
the amount you can bet is limited by how much you have in the channel. The amount you can win is limited by how much the server has in your channel.
MF
10:52
Mr Flintstone
I mean the amount that can be bet by all of the participants
Z
10:52
Zack
I will probably only put like 4 Veo on the server.
10:53
so if you made a channel that had 8 Veo, and the server had 4 of the veo in that channel, then it would block anyone else from making channels. because the server has no more money to put into more channels.
MF
10:53
Mr Flintstone
What exactly is a server ?
Z
10:53
Zack
a full node running amoveo.
10:54
you make channels from the light wallet.
MF
10:54
Mr Flintstone
I think I need to learn more about how channels work
Z
10:54
Zack
you haven't heard of the lightning network?
MF
10:55
Mr Flintstone
yeah, but I do not have a good understanding of how it works exactly
P|
10:55
Paul N. | BAND (πŸ…‘) - I'll never DM you
I think we probably need bounty rewards for tools that aren't directly putting Veo in developer's hands as well. For example, making REST API (if it doesn't already exist) or write exhausive documentaions.
MF
10:55
Mr Flintstone
I thought I had a decent understanding but apparently I need to learn more
Z
10:56
Zack
In reply to this message
Lots of people talk about "REST API", but all the explanations of what that means seem crazy complicated.

Amoveo uses HTTP for its api.
10:56
I haven't seen an explanation of REST API that I can understand.
P|
10:57
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
yeah, and nobody stick to the standard/explanation they set even.
Z
10:58
Zack
the current api is fine for now. It uses JSON.
We can access it using curl, python, javascript, and erlang.
P|
10:59
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
ok, but my point is that there are other things that needs to be developed by people that aren't getting the developer rewards.
11:00
and don't run mining pool
Z
11:01
Zack
People will run markets to make a profit.
11:02
If someone wrote exhaustive documentation right now, it would probably be more trouble than benefit.
It isn't good to duplicate things into the docs if they might change.
Better to only document the parts that people actually use.
11:03
Probably it would be more valuable to search through the docs and find old things that should be deleted.
P|
11:03
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
I can't see how that is related to development of new things that amoveo need, and it needs more man power
Z
11:03
Zack
What things do you think need more man power?
11:04
9 women can't make a baby in 1 month.
11:04
Throwing people at the project doesn't necessarily help.
P|
11:04
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
I am not suggesting what should developers do. I'm suggesting a bounty reward pool for whatever development that's needed out side of your capacity.
Z
11:04
Zack
What development is that?
P|
11:05
Paul N. | BAND (πŸ…‘) - I'll never DM you
Thinga like MEW
Z
11:05
Zack
It seems like you are trying to solve a problem that doesn't exist?
P|
11:05
Paul N. | BAND (πŸ…‘) - I'll never DM you
User interface that allows broader audience to see what's going on in Amoveo
Z
11:06
Zack
I haven't used MEW before. What does it do that you want Amoveo to do?
11:06
The light wallet and explorer are already allowing a broad audience to see what is going on.
P|
11:06
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
But it's not good enough for average joe, UX wise
11:07
The tools need to be more user-friendly, though good UI/UX of necessary.
11:07
I know it's not the "hard" tech, but necessary nonetheless imo
Z
11:08
Zack
sure, we could make an insured crowdfunding contract to pay someone to add more instructions or something to the light node.
11:08
or maybe splitting the page into tabs.
P|
11:10
Paul N. | BAND (πŸ…‘) - I'll never DM you
sorry for not suggesting something very concrete
MF
11:12
Mr Flintstone
In reply to this message
I think the average person who knows how to use MEW should be able to use the light wallet, provided the node it is attached to isnstable
11:12
it is pretty straight forward
11:13
Aside from the whole create/spend thing, but that should be somewhat easily fixable in the light wallet
P|
11:14
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
it is not. I spent some time trying to figure out how to create a new key and download the private key.
Z
11:15
Zack
In reply to this message
Please write down the steps you took, and what was confusing.
That is valuable information for making improvements.
SS
11:19
S S
Zack what's the use of exponent method? Is it better performing than math:pow?
11:51
Deleted Account
is that wise to buy GPU card for mining now?
11:52
i have too small hashrate
MF
12:27
Mr Flintstone
hash rate picking up, 3.8
M
13:26
Mike
In reply to this message
1 VEO
13:27
Any status updates on an AMD miner?
apo joined group by link from Group
P|
15:10
Paul N. | BAND (πŸ…‘) - I'll never DM you
@Mandelhoff Congratulations again for your miner!
15:11
πŸ‘πŸ‘πŸ‘
15:15
Deleted Account
My account shows NaN in the explorer where it has for several days shown a balance. Anyone else having this problem? I'm mining on Mandelhoff's pool
15:18
Deleted Account
yeah, same problem
15:18
seems like Zack's explorer is behind? http://159.65.120.84:8080/explorer.html
15:18
says height 6786
15:18
pool says 6791
15:19
my node also won't sync... again... not even from 0 this time
15:30
pool hasn't found a block for 19 in a row? wonder if network is still alive heh
LB
15:35
Luke B
Is Kassel back up?
15:35
BKyLq is winning every block
J
15:36
Jim
In reply to this message
Could be a sharding problem, let zack know
15:38
Deleted Account
I'm able to sync a node and api:balance() does show a balance for my account
15:39
There seems to be a problem with @Zack's node. It's stuck at height 6786
15:40
^^^ Zack
T
16:09
TownAlwaysGet
Mandel's pool down
I
16:32
Iridescence
Hello, I have updated my Ubuntu CUDA miner to have an auto-tuning parameter tool! Check it out. https://github.com/decryptoed/amoveo-cuda-miner/blob/master/docs/tuning.md
16:32
Also, I have improved my optimized CUDA miner and now see 2.7 GH/s on my GTX1080
16:33
If you missed my previous update, my CUDA miner is also compatible with Mandel's pool
N
16:33
NM$L
Mandel's pool down
S
16:42
Sy
morning....so we forked again?
LB
16:42
Luke B
Mandel pool definitely down
S
16:45
Sy
resyncing...
16:46
6810?
16:49
In reply to this message
that should be amoveopool afaik
LB
16:50
Luke B
amoveopool seems to be down
S
16:50
Sy
yeah, guess he forked or stopped syncing
G
16:52
Gonzalo
my 2 nodes lost sync during the night
S
16:54
Sy
same i guess
16:54
resynced
current height: 6811
terahashes per block: 340
gigahashes per second: 1180
seconds per block: 288
16:55
and restarted my miners, they finally crashed 😝
G
17:03
Gonzalo
17:03
17:04
lots of nodes out of sync :(
S
17:15
Sy
only 4, doesnt seem so bad
G
17:18
Gonzalo
In reply to this message
wrong, light wallet will show you headers height, this will almost be ok, but *explorer* will show you current block height on the node... this is what you have to look at
LB
17:18
Luke B
light wallet will not load for me
G
17:19
Gonzalo
In reply to this message
so I count 17 out of sync :(
S
17:20
Sy
ah okay, only checked the first picture
17:21
they are sleeping right now i guess, i resynced asap when i got up
G
17:23
Gonzalo
In reply to this message
yes I think so, people here on "Amoveo world" is quite tired lately πŸ˜‚
S
17:23
Sy
different timezones too
17:23
its 10 am over her
17:23
e
G
17:23
Gonzalo
we are all out of sync with the world ;)
S
17:23
Sy
😎
G
17:23
Gonzalo
yeah, 10am here as well
17:25
2 days ago I slept from 6am to 7am, and then a 3h nap, so quite outsynced
17:30
hmm
G
17:32
Gonzalo
6810 is Mandels :(
17:32
Mandel wake up man!
17:37
Deleted Account
is this a rewrite or forked from a stable coin?
17:38
Deleted Account
you are talking about amoveo?
17:39
amoveo is totally new coin
Rud C joined group by link from Group
G
17:41
Gonzalo
In reply to this message
amoveo?? no way
17:42
Deleted Account
I wasn't sure if it was based on aeternity or a complete rewrite πŸ™‚
G
17:42
Gonzalo
aeternity could be a bad rewrite, maybe ;)
17:43
Deleted Account
aeternity core was basically made by Zach alone I think
17:43
Deleted Account
so does anyone have a synced node? mine is porked, pool is borked, think I just spent hours mining nothing
17:43
Deleted Account
and he left
17:43
Deleted Account
porked... also borked πŸ™‚
17:44
Deleted Account
have to wait Zach...
G
17:44
Gonzalo
In reply to this message
at this moment amoveo is still experimental software, Zack already said this
17:44
so don't invest to much
17:44
Deleted Account
just asking to see if everyone is having issues, or it is just me, not complaining πŸ˜›
17:44
sounds like the network is dead right now, unti the founder and/or pool operator wake up?
17:45
or am I just doing something wrong?
G
17:45
Gonzalo
In reply to this message
πŸ‘ everybody is having issues, dont care too much
17:45
you can try doing a make prod-clean, you will prob sync again
17:46
Deleted Account
nod, I have tried that, still have 0 blocks; although 6815 headers
G
17:46
Gonzalo
sync:start().
17:46
Deleted Account
I lie, starting to come through now
G
17:46
Gonzalo
great!
17:46
Deleted Account
about the 4th time I've done this - any idea if you can fix it without resyncing the whole blockchain?
17:47
takes quite a long time (for me at least)
17:47
I know it is syncing now my laptop fans are spinning like crazy πŸ˜„
G
17:47
Gonzalo
no idea, sorry
Z
17:48
Zack
im on 6815. Sorry the network was down for a while.
BB
17:48
Brian Brian
Don’t be sorry buddy. You’re not the network you just created it
17:49
Deleted Account
Zack is it a problem that requires some kind of fix/upgrade, or are we just lacking in number of well-connected nodes? maybe we can help with the latter
17:49
I'll reiterate that I'm not complaining, just trying to understand... it makes sense that those of us awake should try to do something if we can πŸ™‚
Deleted joined group by link from Group
Z
17:50
Zack
I think there is some race condition in the software that makes it crash, expecially when blocks are coming quick.
17:50
probably I should combine the tx processing and block processing into a single thread. It is too hard to make them happen in parallel without problems.
17:51
Deleted Account
ahh, I see... I have certainly noticed it works fine for a while then stops syncing and I have to keep telling it to get headers from a peer (per the example command docs)...
17:51
but eventually that stops working and I have to make prod-clean as Gonzalo suggested
17:52
so it seems like you have a solid sync base, maybe this idea will help it stay happy for longer / indefinitely πŸ™‚
Z
17:52
Zack
In reply to this message
sounds like you don't have externally accessible IP address
17:52
Deleted Account
ah yes, that is true, but not something I can fix for my local node
17:52
I was thinking I should run one on a VPS/VM
17:53
it has been "good enough" except for being (eventually) unable to sync blocks and having to restart
Z
17:53
Zack
what you describe is normal intended behavior if you dont have an externally accessible IP address.
No one can send you headers automatically, so you have to pull them manually.
17:54
Deleted Account
oh right, thanks for clarifying that - makes sense now
17:54
people refer to the "light node" or wallet - what is that? maybe I should be running that?
17:55
Deleted Account
ahh, I see
G
18:03
Gonzalo
In reply to this message
yours will be at localhost:8080...
18:04
Deleted Account
oh, so could I access an explorer too?
G
18:06
Gonzalo
What's the best AMD miner at the moment? H/S expected for a rx-580? @Iridescence @Mandelhoff ??
18:08
Deleted Account
Zack thanks I have been using that, I was wondering if I can access it local like the wallet.html, but it seems not?
Z
18:09
Zack
I don't understand the question.
How do you want the explorer to be different?
18:11
Deleted Account
oh just that Gonzalo pointed out I can access the web wallet at http://localhost:8080/wallet.html
18:11
I then thought maybe the explorer is the same, but I guess that's a different thing
Z
18:12
Zack
yes, just replace my ip address with "localhost".
This only works when you are running a full node.
18:12
Deleted Account
oh right, it works, I must have screwed it up before πŸ™‚
18:13
thanks... and sorry for asking a lot of simple questions... I have read through the docs which are pretty decent considering the age and team size πŸ™‚
Z
18:13
Zack
Simple questions show me where to make improvements. I wish more people were asking them.
F
18:25
Francesco @ Simply
You once said that market servers might require gambling licenses. Do full nodes have the option to toggle markets on/off because of this?
18:26
Deleted Account
any plan to hire someone that work for amoveo team?
Z
18:29
Zack
In reply to this message
I said that? I do not recall.
I have no idea what the legal requirement is in your country, I don't know where you are from.

Markets have to be created on a node, and they are not free. You can't accidentally make one.
18:31
In reply to this message
No. Amoveo is a decentralized project which aims to destroy hierarchies like the employee-employer relationship.
It is better to use insured crowdfunding to pay people to develop for Amoveo.
Deleted joined group by link from Group
F
18:33
Francesco @ Simply
In reply to this message
It was more of a thought I guess you posted to the subreddit https://www.reddit.com/r/Amoveo/comments/7rekxw/legal_requirement_to_run_an_amoveo_market
Z
18:35
Zack
oh yeah. It looks like you should put your server in one of those countries I listed.
18:35
if you want to run a market.
F
18:40
Francesco @ Simply
I need to read more about markets, but I like the concept shown in the augur beta. A site where users can create markets and vote on others.
18:41
*bet
Z
18:49
Zack
There is a big issue with the current hashing algorithm.
I think we should do a hard fork as soon as possible.

We only have 2 bytes of nonce to work with, not 7.

This means that the ASIC would have to do 2 blocks of sha256 occasionally, which makes the ASIC a lot more complicated.
18:50
This also makes the GPU more complicated, because each kernel needs to do 60,000 hashes instead of just one.
G
18:52
Gonzalo
nice you found this soon
18:52
tell us what to do and when πŸ‘
Z
18:52
Zack
I didn't find it. Someone emailed me.
G
18:53
Gonzalo
nice, great community
19:15
Deleted Account
Zack, are you suggesting the PoW input will be reduced by 2 bytes from 66 to 64, to avoid a second iteration of SHA-256?
19:17
right now it seems to be [previous block (32), difficulty (2), nonce (32)]
19:17
"We only have 2 bytes of nonce to work with, not 7."
19:17
not quite sure what you mean by that?
19:18
don't you currently have 32 bytes to work with, although you could use 30 and save an iteration, albeit with hard fork?
Z
19:18
Zack
no, we are going down to 55 bytes.
32 for hash of the header, 23 for nonce.
19:18
Deleted Account
will the difficulty not be included any longer?
Z
19:18
Zack
no.
19:19
I will write C code to show how to do the new POW, and then I will give several days for everyone to update their GPU code before we do the hard fork.
19:19
Deleted Account
ok
19:19
one thing that confused me a little, compared to other coins
19:19
is there is no merkle root in the pow hash input
19:20
is the first 32 bytes of the header the previous block hash, or something else?
Z
19:21
Zack
32 bytes are for the hash of the header.
19:22
Deleted Account
right, which presumably includes the previos block hash, merkle root, difficulty, and everything else needed to verify a block πŸ™‚
19:22
if you want to be ASIC/hardware-friendly you might consider aligning + sizing the nonce to a power of two bytes
19:23
In reply to this message
so then, we should do the nonce first, and hash second.
19:23
Deleted Account
was more thinking 7 bytes is an odd number
Z
19:24
Zack
I don't know where you got 7 bytes from.
19:24
23 bytes for nonce, 32 bytes for the hash of the header.
19:24
Deleted Account
oh right, yes you corrected me before on that, sorry πŸ™‚
19:24
23 bytes - 16 24 or 32 would be preferable, I think, but not critical of course
Z
19:25
Zack
you can use 16 bytes of the nonce and ignore the other 7.
19:25
Deleted Account
as it is right now, it's a bit of a pain that the nonce does not line up on 32 (or 64) bit boundaries
19:25
good point
19:25
it is a biiiiig nonce πŸ™‚
19:25
I don't think an ASIC would care but for GPU it's fiddly
19:26
but as you say, can just ignore some of the nonce
19:26
at the moment, there's only 2 bytes which is a bit small given the speed of GPUs (assuming you precompute the first iteration) - so this change would certainly be more efficient πŸ™‚
19:27
even 4 bytes doesn't last long at 3GH/s lol
19:28
32 + 23 works well I guess 32 + 16 + 7 giving a huge nonce space to the miner (16) while allowing pools etc. to have a similarly huge space to distribute work over
19:28
πŸ‘
Z
19:37
Zack
maybe we should do 16 of nonce + 32 of hash + 7 of nonce
That way more things line up at normal boundaries
19:39
Deleted Account
is that really any different from 32+16+7 though ?
19:40
I think it makes the hash not be aligned to 32 or 64 so maybe not preferable (unless it's backwards - endianess?)
19:42
oh sorry I'm getting confused bytes vs bits
P
19:47
Pan
Is there gpu miner for WIN?
19:50
Get the binaries from the releases on that link.
Z
19:55
Zack
In reply to this message
Yes. That is even simpler.
P
20:43
Pan
In reply to this message
Thank you.
Can we use GPU miner in Windows 7?
20:46
Is there a link point to suitable CUDA 9.1 for windows 7?
I just found CUDA Toolkit 9.1 for windows 7.
Does CUDA 9.1 mean CUDA Toolkit 9.1?
P|
20:47
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
yes
P
20:52
Pan
πŸ‘Œ
MH
20:53
Mandel Hoff
You will need Visual Studio 2015 too. It's here:
https://www.visualstudio.com/vs/older-downloads/
20:53
Community Edition is free.
P
20:57
Pan
πŸ‘
N
21:12
NM$L
What's net hashrate now?
Z
21:12
Zack
In reply to this message
N
21:16
NM$L
If i have 2G hashrate, How to calculate how many veo I can get per day?
S
21:40
Sy
terahashes per block: 340
21:41
170000 seconds or 47h to find a block on avg -> 0.5 per day avg
N
21:43
NM$L
In reply to this message
Yes
21:52
In reply to this message
I nearly mine 1 veo today.
OK
21:57
O K
In reply to this message
What is the price?
G
22:31
Gonzalo
+1 (want to know as well)
22:40
Deleted Account
In reply to this message
what's your GPU
N
22:40
NM$L
In reply to this message
1080+1060
22:41
Deleted Account
2g in total?
P
22:42
Pan
Is there a single binary for Windows OS without depending on external environment?
N
22:44
NM$L
In reply to this message
1.6G
P|
22:45
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
That doesn't sound right. With Mandel's miner you should get at least ~2.5 GH/s
N
22:46
NM$L
In reply to this message
P|
22:47
Paul N. | BAND (πŸ…‘) - I'll never DM you
In reply to this message
yes, but you have to use custom numBlock and blockSize parameters
22:49
Deleted Account
do i need to buy a 1080
?
22:50
any possiblity that AMD GPU is more powerful
G
22:50
Gonzalo
??
G
22:50
Gonzalo
G
Gonzalo 10.03.2018 18:06:57
What's the best AMD miner at the moment? H/S expected for a rx-580? @Iridescence @Mandelhoff ??
MH
22:52
Mandel Hoff
Sorry I have no idea about amd. This is my first gpu app and it's very Cuda specific
N
22:53
NM$L
In reply to this message
update. It better now. I get 870M at 1060
S
22:56
Sy
@Mandelhoff you own amoveopool.com right? Can you add a timestamp to your transactions overview?
N
23:05
NM$L
OK
23:20
O K
In reply to this message
I'm not aware of publicly available AMD miner
23:21
Who has paid for Iridescence miner, how much were you charged?
N
23:44
NM$L
1080+1060 = 2.4GH/s Not bad.
G
23:49
Gonzalo
In reply to this message
πŸ‘πŸ‘
11 March 2018
01:27
Deleted Account
ELI5: Why is it better to have ASIC mining when it will almost surely reduce decentralization? Is the fundamental reason that we want to make it costly for miners to switch from mining veo to mining another coin?
I
01:43
Iridescence
I have been asking for donations of 0.1 BTC or 1.2 ETH for my upgrade
O
01:43
Oliver
What is the global hash rate currently?
01:48
Deleted Account
In reply to this message
click Update Height on http://159.65.120.84:8080/explorer.html. Gives the average hashrate of the last 20 blocks
O
01:53
Oliver
So above 2 TH already. Wow
I
01:54
Iridescence
My optimized miner gets 2.6 GH/s on a 1080, 6.3 GH/s on a V100.
Z
01:56
Zack
In reply to this message
Thank you for the community support occamsblade
LB
02:02
Luke B
Am also curious about the switch to ASIC miners
Z
02:05
Zack
In reply to this message
I disagree with your assessment that ASICS will reduce decentralization.
Yes, it is important that Amoveo miners cannot be used on other blockchains.
It is also important that the community can calculate the cost of mining a coin, and it is important that this cost is nearly the same for all miners.
S
02:07
Sy
How is the cost different now? The coin is one week old, it will even out
02:08
Asics will Just make it more exclusive
02:08
Closed source miners will always happen no matter the algo
02:08
In reply to this message
Win/Linux? And/Nvidia?
Z
02:10
Zack
The most important goal when selecting a mining algorithm is that the marginal profit per coin produced is as low as possible. Lower marginal profit = more decentralized.
S
02:10
Sy
Isn't that more determined by diff and price than algo?
02:10
Deleted Account
total net hashrate?
Z
02:11
Zack
In order for the profit margin to be low, it must be an exclusive thing. Only the small fraction of people who are most efficient should be able to mine.
S
02:11
Sy
But I don't think it will work that way, low profits just mean that ppl will rather mine something else
02:12
Hmm interesting but if you got few most efficient miners you aren't really going for decentralisation or am I missing something?
G
02:12
Gonzalo
In reply to this message
well, when this happens (miners go away) -> diif goes down -> margin go up -> start again...
Z
02:12
Zack
If the miner needs to sell 98% of each token just to afford to keep mining, that situation would be very decentralized.
S
02:13
Sy
No...they will all just stop
02:13
You don't mine anything with 2% profit
Z
02:13
Zack
what is the profit margin on bitcoin miners?
S
02:14
Sy
I mine with at least 50% or switch
02:14
Btc dunno, I am in no area where ASIC mining is viable Vs power
02:14
Germany got expensive power
02:15
Plus importing Asics is a headache
Z
02:15
Zack
If Amoveo ASICS can't be used for mining other blockchains, then you can't switch so easily.
02:15
I am not expecting ASICS to exist in a country like Germany which has expensive electricity
S
02:17
Sy
When you go to a non ASIC algo ASICS won't happen for a long time anyway
02:18
Unless you hit your 50% of all coins market cap very fast πŸ˜‚
P|
02:18
Paul N. | BAND (πŸ…‘) - I'll never DM you
non ASIC algo SHA256?
02:18
Deleted Account
In reply to this message
I'm confused. When you say "decentralization" are you talking about decentralization of coin holders or hashpower?
S
02:18
Sy
When you switch to pure btc algo then this spikes we head were northing
02:19
Plus with that slow diff adjustment one farm can essentially destroy mining
02:19
Kassel was nothing compared to that, why would that be desirable?
02:20
We are hitting you with maybe 200 GPUs right now and you want to change algo...going ASIC isn't really the solution
02:20
In a few weeks top everyone is mining with the same speed, just give it time
Z
02:21
Zack
In reply to this message
Neither. I am talking about something more fundamental.

If a miner has 90% profit on their investment, then they can afford to waste 90% of their mining time, and still make a profit.
So they could spend 90% of their time attacking the network without losing money.

If a miner has only 2% profit, then they can only afford to waste 2% of their mining time on attacks, otherwise they are losing money.
02:22
In reply to this message
I don't understand. what is the situation that you think we should avoid?
S
02:23
Sy
What algo do you want to hardfork to?
Z
02:25
Zack
Right now we do work on (32 bytes of the hash of the header) + (2 bytes of difficulty) + (32 bytes of nonce).

We are switching so we instead do work on (32 bytes of the hash of the header) + (23 bytes of nonce.)
I
02:25
Iridescence
In reply to this message
Linux Nvidia
S
02:25
Sy
Thanks @Iridescence
02:26
Asics Is one big centralisation game played by the big players
02:27
You should stay gpu as long as possible if you want to stay decentralised and enable everyone to mine veo
Z
02:28
Zack
In reply to this message
enabling everyone to mine is contradictory with the goal of decentralization.
02:28
I think you are just saying these things because you want to be able to keep mining Amoveo profitably.
S
02:29
Sy
You aren't as profitable as you probably think so no, that's not the reason
Z
02:29
Zack
In reply to this message
S
02:30
Sy
But you don't think as miners do, your 2% margin might be a nice idealistic idea but that's not how it works, none here will mine for 2% unless they have no choice
02:31
And if you go some existing ASIC algo that means that there are other coins to mine as well plus you can be hit with hashrates that are way higher spikes than what we got now
Z
02:32
Zack
In reply to this message
I never suggested reusing an existing ASIC. in fact, I said the opposite.
02:32
In reply to this message
I am not expecting anyone here to be able to profitably mine long-term.
S
02:32
Sy
Yep I noticed, only price blend is, there won't be an ASIC then xD
02:32
Problem
02:33
Gotta go, let's continue this later 😁
Z
02:33
Zack
I will manufacture them myself if no one else does.
S
02:33
Sy
So you got a spare million? Okay
Z
02:33
Zack
There are many types of ASICS. the cheap ones are within my price range.
02:34
We can share a single sheet with other people making ASICS. so the board gets cut apart, and only part of the board is Amoveo ASICS.
02:37
Our hash algorithm is about 4x less complicated than bitcoin's. So our ASICS will be about 4x less costly to manufacture.
S
02:37
Sy
But who will buy an ASIC that makes 2% profit? πŸ˜… Plus the first ones will gain the most again just as we do now
RT
02:40
Roger That
@zack - if you do a simple experiment you will know what the community prefers and what the community prefers to mine will have a better decentralization as more folks will be there or want to be involved - start an amoveo fork with a gpu or cpu only algo and see which one gets more users / miners fast.
Z
02:40
Zack
I don't see why 2% is a limitation?
The gross profit of a miner doesn't have to relate the profit margin.
It is possible to have 50x more money after a year, even if your profit margin per cost of ingredients is only 2%.
02:43
If you reinvest in yourself and make 2% profit every hour, there are 24 * 365 hours in a year, then at the end of the year you will have 2*10^75 times more money than you started with.
That is approaching the number of atoms in the universe.
RT
02:45
Roger That
Profits are calculated annually or over a defined time period. So by default when you say 2% it’s assumed that number is yearly.
02:46
@zack - also I think you believe in ASICS and therefore using that assumption as default in your reasonings. Suggest you drop that. And start the mental model without a bias.
02:47
What do you mean by decentralization? How would you ideally define it using parameters?
02:47
Is it more coin holders, more users, more market cap or more nodes?
Z
02:47
Zack
In reply to this message
@RogerFun I don't see how I can be more clear than this.
02:47
Obviously I am not suggesting 2% profit per year.
02:48
If Amoveo only grew 2% per year, it would take more than my lifetime for it to be worth a significant amount.
RT
02:48
Roger That
Ok let’s start from the beginning. What’s the preferred parameter to you to measure decentralization? More than 1 is fine.
02:49
growth and profit margins are different but lets pause that and do this simple exercise.
02:49
what're the preferred parameters to you?
Z
02:49
Zack
I don't care how "decentralization" is defined.
Efficiency of mining is defined as (value of coins produced) / (cost to produce them).
RT
02:50
Roger That
you have to care Zack because finally that will drive your means to achieve goals for this blockchain
02:50
i understand efficiency
02:52
do you want more full nodes? so you want a large mining network or do you want more coin holders/ users?
02:53
there was a report recently that NEO has just 7 verifying nodes all controlled by the same entity, that's not a blockchain, that's a database
Z
02:53
Zack
I want (cost to produce tokens) / (value those tokens can be sold for) to be as near to 1 as possible.
RT
02:53
Roger That
got it but is that the only parameter you want to focus all your decisions on ?
Z
02:54
Zack
when it comes to mining, yes.
RT
02:54
Roger That
that the marginal cost of coin production approaches 0
Z
02:54
Zack
no, that marginal profit approaches 0.
02:54
marginal cost should approach 1.
RT
02:54
Roger That
i am not asking for mining, i am asking for the entire blockchain?
MF
02:54
Mr Flintstone
I think this happens in any industry, except the marginal profit approaches cost of capital
RT
02:54
Roger That
sorry yes
MF
02:54
Mr Flintstone
not zero
RT
02:54
Roger That
1
02:55
i believe that happens itself through competition, you dont have to force the design to achieve it
Z
02:56
Zack
the natural trend of competition is in this direction yes.
So we just need to encourage the natural tendancy.
RT
02:56
Roger That
if you have a very popular commdity that can be easily (technologically easily) produced then it always achieves that curve
02:56
so when then try to achieve something which is anyways achieved by default?
Z
02:57
Zack
If we used CPU, then the curve would look very different. Botnets don't follow the curve very closely.
The botnets fight each other for control, and whoever controls the dominant botnet could produce coins at almost no cost.
RT
02:58
Roger That
that was true for bitcoin botnets (if any - i am not sure) before asics came along
02:58
so competition itself took care of it
Z
02:59
Zack
If ASICS were impossible, and only CPU mining as possible, then Bitcoin would have never recovered. it would have been a fatal flaw.
RT
02:59
Roger That
but i doubt if satoshi designed it such way so that asics will be born
Z
02:59
Zack
so we can't just ignore this problem and expect it to solve itself.
RT
02:59
Roger That
it played out that way
Z
02:59
Zack
It doesn't matter what was going on in Satoshi's head.
MF
02:59
Mr Flintstone
In reply to this message
i see your argument now. Despite market forces trending toward low marginal profit, we still need to be mindful of design decisions here
RT
03:00
Roger That
it matters because today marginal cost of bitcoin mining is close to 1 and not because satoshi had to forcefit that parameter, it evolved like it always does
MF
03:01
Mr Flintstone
In reply to this message
if they had accidentally chosen a CPU only algorithm then that may not of happened
RT
03:01
Roger That
also i also believe that both extremes are dangeous for a blockchain - CPU mining and ASIC
MF
03:01
Mr Flintstone
have* lol
RT
03:01
Roger That
better to find a middle ground to strike a compromise
Z
03:02
Zack
Logical arguments are appreciated here. "beliefs" are not.
RT
03:02
Roger That
compromise of mining cost and decentralization
03:02
In reply to this message
alright then. discussion stops here.
MF
03:03
Mr Flintstone
@zack so if I understand correctly, you want to change PoW algo. In part, this motivation is driven by a desire for cheaper ASICS
Z
03:03
Zack
In reply to this message
yes
03:04
it might also prevent Kassel from having such a big advantage.
MF
03:04
Mr Flintstone
why?
OK
03:05
O K
Where is the evidence that 1) ASIC miners have a lower marginal profit, 2) ASICs reduce centralization
Z
03:05
Zack
right now, if you do the partial hash optimization, then you only have 2 bytes of nonce space.
So there are some potential optimizations that make this into a partially memory-hard problem.
OK
03:05
O K
It's pretty obvious as a miner that 1) GPU have wider distribution 2) lower barrier to entry 3) lower marginal profit
Z
03:06
Zack
In reply to this message
if you don't use an ASIC, then the miners will just keep switching blockchains, and the marginal profit will be high on all the blockchains.
It is a way of attacking the retargeting algorithm. We will have 1 period of 1 hour-long blocks, followed by 1 period of 10-second long blocks.
OK
03:07
O K
The marginal profit is not high on all blockchains
03:07
That's an assumption you're making that is unbacked by evidence
Z
03:08
Zack
It is the Nash equilibrium.
I doubt any evidence you have is accurate anyway.
OK
03:08
O K
I'm asking you for evidence
03:09
My evidence is being a miner, something you've admitted yourself you have little real world experience with
Z
03:09
Zack
https://www.multipool.us/
here it is happening right now.
03:10
oh, maybe not that one.
OK
03:10
O K
Yes, I take no issue that GPU miners move to the chain that is most profitable. I agree, they do. That has nothing to do with *marginal profit*
03:10
"across all chains"
Z
03:11
Zack
when GPU miners can switch to the most profitable chain, then the other chains don't have enough work being done, so they retarget to make the tokens too cheap.
OK
03:11
O K
"enough work" "too cheap"
MF
03:11
Mr Flintstone
In reply to this message
one thing I would add is that if we have a unique asic, the chain should be more resistant to reversion
OK
03:11
O K
It seems to me from my study of economics that they would "just the right amount of work" and they would retarget to make the coins "just the right price"
MF
03:12
Mr Flintstone
at least while it’s small
Z
03:13
Zack
In reply to this message
there is a delay of 2000 blocks before the blockchain reacts to changes in mining power.
Miners take advantage of this delay to always mine the cheapest tokens.
OK
03:13
O K
I'm afraid there is no evidence that you can produce that ASIC miners have lower marginal profits, and there is an inherent contradiction in your values here
Z
03:13
Zack
The math is pretty simple. I am not sure where you are getting confused.
OK
03:14
O K
Show me the evidence, where is the evidence that ASIC miners have lower marginal profits
03:15
The barrier to entry is higher for obtaining ASICs, and now patents are being used to stifle innovation
Z
03:15
Zack
The Nash equilibrium is for GPU miners to keep switching to the most profitable chain.
This means that each blockchain will alternate between having 2000 slow 1-hour blocks, followed by 2000 fast 10-second blocks.
During the 10-second block period, the marginal profit is much higher.
OK
03:16
O K
The evidence doesn't show that https://etherscan.io/chart/blocktime
03:17
In reply to this message
"During the 10 second block period the marginal profit is much higher" what about the average marginal profit between ASIC miners and GPU miners
Z
03:18
Zack
The effect is minimized on the largest blockchain using any particular mining technology.
OK
03:19
O K
Show me where the marginal profit over a year for ASIC miners is lower than GPU mining farms
03:19
doesn't make sense, doesn't exist
Z
03:19
Zack
These are things that we can't accurately measure.
To design a good blockchain you can't be distracted by the news. Instead we need to focus on the game theory.
OK
03:20
O K
It's not news, it's the truth
03:21
You can do the math on buying a GPU right now and mining on any chain, knowing hashrate, estimate difficulty increase, and see the marginal profit
M
03:26
Mike
GPU mining isn’t that decentralized because of pools
S
03:28
Sy
Asics will be bought by big farms too and they will hit whatever they want
OK
03:28
O K
In reply to this message
What do you mean?
S
03:28
Sy
Most btc ASICS are divided between less than 10 players I think
03:29
You keep quoting that you want to prevent Kassel but that's not reality, there will always be someone with more money and thus more hashlower than others
03:29
And it will always amass over time and make single players more powerful
03:29
Bitmaintech runs what? 10-20% oft btc network?
Z
03:29
Zack
Another big advantage of ASICS is miner loyalty. They don't want to destroy the blockchain if it will make their ASIC worthless.
S
03:30
Sy
Why would any gpu miner want to destroy the Blockchain? Plus it's always a matter of the number, if it is big enough some ppl couldn't care less about their initial investment
OK
03:30
O K
In reply to this message
Historically this hasn't played out as as big a danger as was feared in the early days
s
03:30
scott_l | DeFi Pulse
@zack, what do u think about scalar markets? conditional markets? and then interoperability w other coins?
OK
03:30
O K
In reply to this message
Many miners are also loyal, and with hobbyist miners: set it and forget it types
OK
03:31
O K
You underestimate the human resources that go into changing chains, uptime, and maintenance
RT
03:31
Roger That
they have asics but still working on interests contrary to main blockchain
OK
03:32
O K
From what I understand ASIC boost required empty blocks right?
RT
03:32
Roger That
yes so they worked in maximizing their own benefit which was contrary to benefit of the blockchain
03:33
maybe consider than in game theory
OK
03:33
O K
The fact that a 10 year old girl with an interest in the future and blockchains can turn on her gaming PC and mine her favorite coin regardless of profitability says something
03:34
Patents == intellectual monopoly == barrier to entry == greater marginal profit
03:34
In any industry
RT
03:35
Roger That
anyways, we are speaking on governance matters but i was comfortably shutdown by zack for speaking out on a governance parameter. i used the word "belief" which is what is being used to decide various blockchain parameters here without "explicitly" using the term. so seems standards are inconsistent.
03:36
you guys carry on and reach the nash equilibrium, i am sure i will not be missed here. telegram user count going down from 311 to 310
OK
03:37
O K
Awe man
M
03:38
Minieep21
Has anyone tried mining on Google Cloud? They got P100s as well
OK
03:41
O K
In reply to this message
Thank you for clarification
Z
03:42
Zack
In reply to this message
The smart contracts are turing complete. I hope we program conditional markets soon.
Since the oracles only give us binary data, a scalar market isn't possible. But we can encode a number in binary, and use a different oracle for each bit. So we can program something that works like a scalar market.

interoperability: Amoveo has lightning payments already. atomic swaps are a small variation on this. I hope we get atomic swaps working soon.
03:45
combinatorial markets can probably use nearly the same code as binary markets. Just with 4 outcomes instead of 2.
03:46
Deleted Account
Zack, in the class of attacks you want to protect against, the attacker seeks to profit/break-even from Amoveo while he's attacking it. Three questions:

1) Do you mean a 51% attack? If not, what does it matter if they mount an attack?

2) What about the class of attacks where the attacker is fine with running at a loss? (because, say, he has a personal beef with you and wants to see Amoveo fail.) Here it's not so much about the profit-margin as it is about the total cost of the attack. Isn't quickly building up a large distributed base of honest hashpower the best (only?) defense? Won't this be harder to do with an ASIC-easy algo?

3) Following up on Q2) is the primary worry that gpu-miners will switch to another coin when the difficulty of amoveo is high leaving amoveo vulnerable to a 51% attack?
OK
03:51
O K
In reply to this message
Botnets have always been valuable and have opportunity costs. I've pointed this out before and it's important to remember.
Deleted joined group by link from Group
03:51
Deleted Account
Hi
03:53
What kind of gpus here yield best results, amd or nvidia?
M
03:53
Minieep21
In reply to this message
No AMD miner yet
03:54
Deleted Account
In reply to this message
Ok thank you for the information.
R
03:59
RBT
@Mandelhoff is the pool working?
04:00
i just noticed i am getting an error. GetWork failed. Sleep and retry...
04:00
Deleted Account
What’s the development team size here?
Z
04:00
Zack
In reply to this message
Well, I talked about a few different kinds of problems recently.
1) when miners keep switching blockchains, and the difficulty alternates high-low
2) when it is too cheap to produce tokens, and miners profit in a way that harms everyone else. (this isn't exactly an "attack", but it is a problem)
3) 51% attacks

It is impossible to make a blockchain secure against an attacker with infinite resources to burn. So instead we try to make it as expensive as possible.

So lets look at how 51% attacks work in GPU vs ASICs.

Having more hashpower is important for this it is true, but you need to think of "more" as in a larger percentage of the world's hashpower of this type of hashing algorithm, and you need to think of it as value per hash.

For instance, a bitcoin hash currently costs something like 200 000x less than an Amoveo hash.
So having 4 TH/s on Amoveo is more than having 10 000 TH/s on bitcoin.

If only 10% of the worlds GPU are pointed at your blockchain, then it isn't so hard for the attacker to convince 10% of GPU owners to sell their power to him. Since the market is so big, it will be liquid. The price will stay about the same as he buys more hashpower.

If 90% of the worlds asics of this type are pointed at your blockchain, then it is more difficult for the attacker. The market is much less liquid, so the price rises quickly as he purchases more hashpower.
04:02
Deleted Account
So this can be mined with ASICs??
OK
04:02
O K
In reply to this message
1 and a big annoying community
Z
04:02
Zack
In reply to this message
We are all on the development team, though most of us are testers.
S
04:02
Sy
2 will never happen for long, it would be nice otherwise
04:03
If you got big profits on a coin it attracts others and just equals out to normal
OK
04:04
O K
In reply to this message
πŸ‘
04:04
Deleted Account
In reply to this message
Thanks. So bitcoin ASICs already mining this?
04:05
Fuck me. How old is this network ?
S
04:05
Sy
No
04:05
One week
OK
04:05
O K
In reply to this message
I'm kidding, Zack is the one busting his ass. No ASIC, no public AMD miner yet
S
04:06
Sy
Own ASIC would only work if it really worth it, otherwise most will just skip the net
04:06
Deleted Account
Ok I am still reading older posts.
S
04:07
Sy
You can control the hashrate distribution a bit though until it is really worth it and bitmaintech or so designs there own
04:08
And then some farm with a few thousand units just kills every small miner 😭
04:08
Deleted Account
Ok so gpus still have some time here it seems.
04:08
Why is the roger guy upset?
S
04:08
Sy
Dunno
04:08
Deleted Account
Sorry newbie here. Too many questions.
D
04:08
Denis
Looks like @Mandelhoff pool down
OK
04:09
O K
No need to get into community politics so soon, stick around and let them develop arround you Terrible
04:09
Welcome!
04:09
Deleted Account
Ok. Yeah good idea.
04:10
I am not planning to get involved in decisions anyways.
04:10
Just unearth some hashes πŸ˜‚
OK
04:10
O K
@Mandelhoff pool down for me too @cryptohunter77
D
04:11
Denis
In reply to this message
It's back to online nowπŸ‘
Z
04:12
Zack
In reply to this message
Yes it can. If the difficulty oscillates up and down periodically, and miners keep switching to the more profitable blockchain.

It is kind of like when you are draining a bottle. It goes a lot faster if you make the water rotate.
OK
04:13
O K
I prefer just squeezing the bottle
04:13
🐳
S
04:14
Sy
hmm another bad black? my wallet is going crazy again
04:14
Deleted Account
In reply to this message
Maybe just drill a hole on the back. 🀣
S
04:14
Sy
my node*
OK
04:14
O K
In reply to this message
Haha yeah
S
04:15
Sy
hmm my node stopped syncing again, 20 blocks behind spamming potential bad block
04:16
died died died πŸ˜‚
OK
04:16
O K
haha I always crack up at that too
04:16
I love error messages in general
Z
04:16
Zack
erlang keeps trying to restart processes if they die.
S
04:17
Sy
any reason why i got out of sync? or just shit happens?
04:17
Deleted Account
Fuck, difficult so high! How many gpus do each of you guys have?
S
04:17
Sy
last time it was mandels payout xD not sure about tonight
Z
04:17
Zack
I think there is some race condition between tx processing and block processing.
S
04:17
Sy
and this time, dunno
04:17
okay so just bad luck...only 20 blocks, not that bad
Z
04:17
Zack
I guess the thing to do is to put them both on one thread.
OK
04:17
O K
Terrible, I have only three and I do very well
S
04:17
Sy
in total about 60
OK
04:18
O K
I get 1 veo a night. The block reward is small and there are few Veo
Z
04:18
Zack
ive been thinkning about tx propagation.
It is pretty easy to waste bandwidth here if we have a bad design.
04:18
Deleted Account
In reply to this message
What are you nvidia testing lab?
S
04:18
Sy
semi pro miner
04:18
i would like to live from what i mine...i think i can do that within this year
Z
04:19
Zack
I think a short term solution is that every time you make a tx, it automatically gets pushed to a hard-coded list of the best mining pools.
This way we have time to come up with a more time to build the long term solution.
04:19
Deleted Account
In reply to this message
What’s your monthly power bill like ? Lol 😝
S
04:19
Sy
the new 1070ti are not that bad, i think im at 700€
04:20
i power limit them at 130W-150W though, 180W is only marginaly faster
04:20
Deleted Account
Lol. Alright. You can sell your heat to a small cathedral and make some extra money.
Z
04:20
Zack
And the long term solution is to share a list of checksum values genrated from each tx.
Then you request a list of txs based on their checksum.

It seems to me that this checksum can be 4 bytes, and we don't need collision prevention?
04:21
Deleted Account
How many of you 60 gpu fuckers are here?
S
04:21
Sy
I would love to heat a pool with it tbh πŸ˜‚
04:21
at least two more
Z
04:22
Zack
pool-heated pool.
S
04:22
Sy
its like i said, good profits attract miner, doesnt matter what algo you choose, if you become bigger so will the farms you attract
OK
04:22
O K
xD
S
04:22
Sy
im pretty small tbh
04:22
Deleted Account
Fuck man, you can retire in half of Asia if you just sell of your gpus at market price today.
S
04:23
Sy
lol
04:23
that thought crossed my mind tbh :D i could probably sell my rigs with a nice profit, im running closed 19" cases
M
04:23
Minieep21
Was there a linux miner for Mandel's pool?
04:24
Deleted Account
In reply to this message
Next project - make your own windmill for free electricity.
S
04:24
Sy
i thought about that aswell xD i live in a pretty windy area
04:25
but its a bitch in germany with all those regulations and building permits
04:25
Deleted Account
I think I would rather try to buy after a while.
S
04:25
Sy
around here at any given point with a few you see 40-100 windmills
04:25
Deleted Account
Germany huh. Fucking insane with all those costs and taxes.
S
04:25
Sy
with a view*
04:25
Deleted Account
Is crypto taxed there?
S
04:26
Sy
mining yes, normal income
OK
04:26
O K
@imaik tumblecatweed ported it
04:26
Deleted Account
Get a boat and take it to international waters and mine thereπŸ€ͺπŸ€ͺπŸ€ͺ
S
04:26
Sy
as it is in the US and every "normal" country :D
OK
04:26
O K
In reply to this message
that's the difference between pro and semi-pro xD
04:27
Deleted Account
In reply to this message
Pro is asic I think.
S
04:27
Sy
yep, all my mining is running through a company, all clean and legit :)
OK
04:27
O K
Pro is no taxes
S
04:27
Sy
na, asic is...different
04:27
for the lazy ones
04:27
i like my gpus
04:27
Deleted Account
Pro is asic in China.
S
04:27
Sy
cant find stuff like this with asics ^^
04:27
even in china i would run a gpu farm, profits are way higher
04:27
you wont find a lucky coin with asics and ROI in a week
04:28
with gpus...been there, done that
04:28
Deleted Account
And buying a power plant to mine and fry the neighborhood with radiation and heat.
S
04:28
Sy
i checked watermills in austria, there are a few ^^
04:28
Deleted Account
Lol
S
04:28
Sy
but if you calculate power vs extra time spent and extra logistics...doesnt really matter
04:29
if i pay 200 or 1000€ per month isnt really that much vs your profits, they do scale as you grow :)
OK
04:29
O K
Hey zack, what region are you in?
04:29
Central/south america?
04:29
Deleted Account
Zack is satoshi?
S
04:29
Sy
at some point it might be 10k vs 4k power cost but then again the number at the end of the month will be much bigger too 😎
OK
04:29
O K
In reply to this message
it's not clear
S
04:29
Sy
hehe
OK
04:32
O K
In reply to this message
If only satellites had better latency
S
04:33
Sy
taxes arent that bad tbh, you do pay them on your profits after all so you did make money
Z
04:33
Zack
There are some places in the world with big dams making lots of electricity, but since these locations are so remote, they waste a lot of energy by the time they transport it all the way to population.
So if you build the miner right next to a dam like this, I think it might be best.
I wonder if there is a way to dump computer heat into a river.
S
04:33
Sy
canada just sold such a spot