★★DigiByte|极特币★★[DGB]✔ Core v6.16.5.1 - DigiShield, DigiSpeed, Segwit

[Kayahoga]

Not that I can really compete with 10GH but if we know when all the power is being thrown at the network we can do a coordinated effort to help push things along. 

[DigiByte]

This could be a problem ...

I pulled all my hash from the network around block 142,410 ... it then ran ok for the next 12+ hours until block 142,473 where the difficulty has now suddenly jumped to 1,000+ the highest digibyte has ever seen... could be stuck on this block for a long time.

I could probably clear it, but am getting a bit bored throwing away my own money to push the blockchain along ... especially as the Devs have not come forward and explained the cause of the problem. Happy to support them if they come up with a solution and need hashing power, but until then I have retreated to the side lines.

Also, does anyone know of a way to monitor the DGB network hashrate? All the usual sources no longer work since the introduction of the new wallet.

First off, thank you very much for putting your hash toward the network and trying to help us ll get over the hump. We dont expect anyone to mine for days on end with no return. As for the problem we have mentioned it several time over the past couple months.

The problem is not with DigiShield alone as it worked fine until we combined it with the 0.9.1 protocol in April (you can see this on several charts.) DigiShield with the 0.6 Bitcoin protocol works great for several coins still and it still works for Dogecoin. We screwed up and did not do enough testing when we released the 0.9.1 wallet. We figured since no hard fork was required everything with the difficulty adjustment would stay the same.

What basically happens with the 2.9.1 protocol is when a sudden hash spike occurs blocks are sometimes discovered only milliseconds apart. Combine this with a single block average (DigiShield only goes back to the previous block) and you get an astronomically high inflated difficulty.

The next problem is when the hash rate drops and the difficulty eventually drops it drops too low and an oscillating effect occurs & the process repeats itself.

The best way to overcome this is with a steady increase in hash, not a sudden spike. Also, a consistent hash base will be a huge improvement. Starting tomorrow we will begin adding hash to the network and we will publicly post the days target hash-rate. This will help speed things up considerably. We ask if you have some hash to contribute we could definitely use it.

We dont expect people to do this for weeks on end but if we could get a rotation going that would be great.

@Digibyte

Thanks for the response, much appreciated.

I admit I don't have all the answers but if it is the new protocol that is causing the problem then it is a concern that it will rear its head after the wallet passes block 145,000. The explanation you gave is plausible, but NOT the whole story and it doesn't explain what I have been observing, examples:

A. Block 142,272 - 142,273 took over 5 hours to find the block at a difficulty of 1022 ... by your reckoning Digishield would then reduce the difficulty far to much ... but it actually increased to 1300+

B. Whilst pointing several GHs of hash at the network I was finding several groups of consecutive blocks, looking at these I often found two or more blocks in quick succession (seconds apart) but rather than increasing the difficulty as you suggest it actually reduced it for the next blocks.

C. It is true the difficulty is oscillating, but it is also oscillating at far too high a range. It never reduces the difficulty to a point where you would find a block on average every minute (ie the correct difficulty) ... any idea why it fails to retarget to the correct level? Even when there has been very little hash pointing at the network it still oscillates roughly in the range 200-400.

D. By increasing the network hash rate x6 I thought the difficulty may have gone up ... but it actually stayed the same and still oscillated in that same range.

Do you have any idea why the hashrate has jumped to record levels over the last two blocks? I just want to make sure we don't replicate that and get it permanently stuck at an even higher difficulty.

Interesting there are no problems with other coins using Digishield, Doge you might expect as it maintains a consistent very high hashrate ... but I would have thought some other coins may have had similar problems by now. Could it be unique to Digibyte and a stray line of code that is not included in other coins?

Anyway, hopefully this will all be resolved and we can cruise to block 145k in no time.

Happy to help with the hash power if you set a daily target. Is there somewhere we can check the Digibyte Network hashrate? Most of the original sources seem to have 'broken' since the wallet update.

A) There is actually a delay with the code that offsets things. So if there is a sudden spike the difficulty will not respond to that until the 3rd or 4th block. We witnessed this wile testing the original DigiShield. Also, remember that block discoveries are random and not an exact process. Blocks can "accidentally" be discovered back to back even at a very high difficulty, low diff situation.

B & C) This is an interesting observation we saw after the 2.9.1 codebase integration which we unfortunately did not test or recognize until after the fact.

When we first tested Digishield (v2.0) over the course of a couple weeks we routinely saw the oscillation. In fact the oscillation is still present in Dogecoin. But it was a trade off for resistance against multi-pool attacks. I wish I could point you to an exact line of code in the 2.9.1 codebase but we are positive it was something we unintentionally integrated with the 0.9.1 Bitcoin core update that was meant for the Bitcoin difficulty adjustment scenario.

3.0.1 was built from the ground up from a completely different codebase. The difficulty issue will not be seen with the multi-algo update. Technically speaking the difficulty adjustment will work completely different.

[Scyntech]

I forgot to ask. Every time I close the v3.0.1.0 wallet and bring back up, it will hang for a good deal of time at 1 hour -4 hours behind. I've tried fresh installs and it keeps doing it. Should I go back to an older wallet for the time being?

[NZ_miner]

@Digibyte

Thanks, that makes much more sense now.

Bit of a long shot, but I will drop josh a line and see if he is feeling generous ... failing that I think we should go with a coordinated push, do you have an idea for a start time/date so everyone can get involved?

[DigiByte]

I forgot to ask. Every time I close the v3.0.1.0 wallet and bring back up, it will hang for a good deal of time at 1 hour -4 hours behind. I've tried fresh installs and it keeps doing it. Should I go back to an older wallet for the time being?

What OS are you on?

[Scyntech]

I forgot to ask. Every time I close the v3.0.1.0 wallet and bring back up, it will hang for a good deal of time at 1 hour -4 hours behind. I've tried fresh installs and it keeps doing it. Should I go back to an older wallet for the time being?

What OS are you on?

Windows 7 64

[DigiByte]

@Digibyte

Thanks, that makes much more sense now.

Bit of a long shot, but I will drop josh a line and see if he is feeling generous ... failing that I think we should go with a coordinated push, do you have an idea for a start time/date so everyone can get involved?

How does 6 PM UTC time sound tomorrow? Thats 2 PM Eastern, 11 AM Pacific?

Checking into the hash reporting issue with 3.0.1. According to the digi.morecoins pool (2.9.1 wallet) we are currently at 1121 MH net hash. http://digi.morecoins.org/index.php?page=dashboard

We had an issue with the 3.0.0 wallet where it would crash when you ran "getinfo" or "getmininginfo." This may be related.

[DigiByte]

I forgot to ask. Every time I close the v3.0.1.0 wallet and bring back up, it will hang for a good deal of time at 1 hour -4 hours behind. I've tried fresh installs and it keeps doing it. Should I go back to an older wallet for the time being?

What OS are you on?

Windows 7 64

You downloaded the version 3.0.0 wallet correct?

We had an issue on two of our PC's that we tested the 3.0.0 wallet on & then installed the 3.0.1 wallet. We had to delete the "blockchain" & "chainstate" folders and let the wallet re-download the block chain. 3.0.0 forked for a few hours so chances are your block-chain & database is corrupted in the Appdata folder. This would not be fixed by uninstalling/re-installing the wallet. Think we also deleted may have deleted the database file. Try deleting those files in AppDate/Romaing/DigiByte.

[Scyntech]

I am running the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) which I downloaded from the main site. I tried deleting the "blockchain" & "chainstate" folders. Even with a fresh install (all traces of the digibtye data removed) it still will hang with the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) version. I had no issue prior to this version wallet. If that helps any


I think what I will do for now since I currently only have one ATI 7850 (tiny 345Kh/s) is wait for block 145,000. Then I can add 38Gh/s with my USB miners.

[DigiByte]

Lets target 3GH net-hash at 6PM UTC tomorrow (2 PM Eastern, 11 AM Pacific.)

Current NetHash is: 1121 MH. (Taken from v2.9.1 wallet data)

In the past DigiByte had 2-3 GH routinely before the slow block-chain problem started. The key is to normalize the hash-rate & keep it consistent in order to let the difficulty stabilize.

[DigiByte]

I am running the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) which I downloaded from the main site. I tried deleting the "blockchain" & "chainstate" folders. Even with a fresh install (all traces of the digibtye data removed) it still will hang with the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) version. I had no issue prior to this version wallet. If that helps any


I think what I will do for now since I currently only have one ATI 7850 (tiny 345Kh/s) is wait for block 145,000. Then I can add 38Gh/s with my USB miners.

Interesting. Can you provide a screenshot & debug.log info? We have 5 Windows 7 x64 bit machines running 3.0.1 in the office & at home with no issues.

[Scyntech]

I am running the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) which I downloaded from the main site. I tried deleting the "blockchain" & "chainstate" folders. Even with a fresh install (all traces of the digibtye data removed) it still will hang with the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) version. I had no issue prior to this version wallet. If that helps any


I think what I will do for now since I currently only have one ATI 7850 (tiny 345Kh/s) is wait for block 145,000. Then I can add 38Gh/s with my USB miners.

Interesting. Can you provide a screenshot & debug.log info? We have 5 Windows 7 x64 bit machines running 3.0.1 in the office & at home with no issues.

It started reindexing again =/

EDIT- shut it down and brought it back up and now it's at the 5 hour behind mark again. I tried adding the debug info and the forum won't allow me to post it. Is there a specific part you need to look at?

Edit 2- I think I found the issue... maybe. I went to the main and downloaded the wallet for Windows 64bit. I noticed the file name was different than the version I had before. I may have install a previous wallet version. I installed the current wallet from the site. Hopefully this works.

[NZ_miner]

I am running the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) which I downloaded from the main site. I tried deleting the "blockchain" & "chainstate" folders. Even with a fresh install (all traces of the digibtye data removed) it still will hang with the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) version. I had no issue prior to this version wallet. If that helps any


I think what I will do for now since I currently only have one ATI 7850 (tiny 345Kh/s) is wait for block 145,000. Then I can add 38Gh/s with my USB miners.

Interesting. Can you provide a screenshot & debug.log info? We have 5 Windows 7 x64 bit machines running 3.0.1 in the office & at home with no issues.

This sounds as if it is just the slow block issue ... ie because a block has not been found for 4+ hours the wallet shows out of sync (even though it is up to date) because it has taken so long. If you hover over the information and look at the latest block received and it matches the last one shown in a block explorer all should be good.

I imagine this will not be a problem once we get to block 145,000 and we hopefully return to a block a minute average.

[NZ_miner]

@Digibyte

Thanks, that makes much more sense now.

Bit of a long shot, but I will drop josh a line and see if he is feeling generous ... failing that I think we should go with a coordinated push, do you have an idea for a start time/date so everyone can get involved?

How does 6 PM UTC time sound tomorrow? Thats 2 PM Eastern, 11 AM Pacific?

Checking into the hash reporting issue with 3.0.1. According to the digi.morecoins pool (2.9.1 wallet) we are currently at 1121 MH net hash. http://digi.morecoins.org/index.php?page=dashboard

We had an issue with the 3.0.0 wallet where it would crash when you ran "getinfo" or "getmininginfo." This may be related.

That is good for me, I will start adding hash shortly after 6pm UTC

[Scyntech]

I am running the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) which I downloaded from the main site. I tried deleting the "blockchain" & "chainstate" folders. Even with a fresh install (all traces of the digibtye data removed) it still will hang with the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) version. I had no issue prior to this version wallet. If that helps any


I think what I will do for now since I currently only have one ATI 7850 (tiny 345Kh/s) is wait for block 145,000. Then I can add 38Gh/s with my USB miners.

Interesting. Can you provide a screenshot & debug.log info? We have 5 Windows 7 x64 bit machines running 3.0.1 in the office & at home with no issues.

This sounds as if it is just the slow block issue ... ie because a block has not been found for 4+ hours the wallet shows out of sync (even though it is up to date) because it has taken so long. If you hover over the information and look at the latest block received and it matches the last one shown in a block explorer all should be good.

I imagine this will not be a problem once we get to block 145,000 and we hopefully return to a block a minute average.

I'm hoping that's the issue or I just screwed something up myself rather than it be a wallet issue.

[DigiByte]

I am running the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) which I downloaded from the main site. I tried deleting the "blockchain" & "chainstate" folders. Even with a fresh install (all traces of the digibtye data removed) it still will hang with the v3.0.1.0-g354c0f3-MultiAlgo (64-bit) version. I had no issue prior to this version wallet. If that helps any


I think what I will do for now since I currently only have one ATI 7850 (tiny 345Kh/s) is wait for block 145,000. Then I can add 38Gh/s with my USB miners.

Interesting. Can you provide a screenshot & debug.log info? We have 5 Windows 7 x64 bit machines running 3.0.1 in the office & at home with no issues.

It started reindexing again =/

EDIT- shut it down and brought it back up and now it's at the 5 hour behind mark again. I tried adding the debug info and the forum won't allow me to post it. Is there a specific part you need to look at?

Edit 2- I think I found the issue... maybe. I went to the main and downloaded the wallet for Windows 64bit. I noticed the file name was different than the version I had before. I may have install a previous wallet version. I installed the current wallet from the site. Hopefully this works.

The 3.0.1b wallet only included language translation updates. No changes were made other than that. If your an English speaker you wont notice a difference. When we ported the code we had some pretty wild foreign translations for the word "DigiByte."

[Scyntech]

The 3.0.1b wallet only included language translation updates. No changes were made other than that. If your an English speaker you wont notice a difference. When we ported the code we had some pretty wild foreign translations for the word "DigiByte." Smiley

So far so good. Wallet updated with the older wallet.data file perfectly. I have shut it down and opened it 5 times. Synced up quickly

[tdcooper99]

Nice to see difficulty has dropped again from 1300+ back down to 316. But still too high for the current block discovery rate...  but hopefully we can get things coordinated for this evening. 6PM UTC...  for us in the EU we're on summer time now so it's actually 7PM GMT and 8PM for Western Europe.

Hopefully we'll get some help from Josh @ Gawminers.

[munky87]

As some people ask, how can we now get the scrypt hashrate ? I think it is really important to coordinate efforts.

[appbox]

As some people ask, how can we now get the scrypt hashrate ? I think it is really important to coordinate efforts.

https://www.nicehash.com/index.jsp

https://www.westhash.com/index.jsp